Available via license: CC BY 4.0
Content may be subject to copyright.
342
I
nt. J. Risk Assessment and Management, Vol. 22, Nos. 3/4, 2019
Copyright © The Author(s) 2019. Published by Inderscience Publishers Ltd. This is an Open Access Article
distributed under the CC BY license. (http://creativecommons.org/licenses/by/4.0/)
Improving decision making for carbon management
initiatives
Victoria Campbell-Árvai*
School for Environment and Sustainability,
University of Michigan,
440 Church Street,
Ann Arbor, MI 48109, USA
Email: vcarvai@umich.edu
*Corresponding author
Douglas Bessette
Department of Community Sustainability,
Michigan State University,
480 Wilson Road,
East Lansing, MI 48824, USA
Email: bessett6@anr.msu.edu
Lisa Kenney
Greater Buffalo Niagara Regional Transportation Council,
438 Main Street, No. 503,
Buffalo, NY 14202, USA
Email: lkenney@gbnrtc.org
Joseph Árvai
School for Environment and Sustainability,
University of Michigan,
440 Church Street,
Ann Arbor, MI 48109, USA
and
Stephen M. Ross School of Business,
University of Michigan,
701 Tappan Street,
Ann Arbor, MI, 48109, USA
and
Decision Research,
1201 Oak Street,
Eugene, OR 97401, USA
Email: jlarvai@umich.edu
I
mproving decision making for carbon management initiatives 343
Abstract: This paper reviews five challenges faced during decision making
about carbon management initiatives. The first of these challenges deals with
behavioural and perceptual obstacles, which often leads to the introduction of
systematic biases during decision making. The remaining four obstacles deal
with the complexity associated with the carbon management problems
themselves. These include neglecting the objectives and related measurement
criteria, which will guide decisions among competing risk management
options; the tendency to look for singular solutions to complex problems, rather
than considering a broad array of options; a lack of explicit attention devoted to
the full range of tradeoffs that should be considered when choosing among
alternatives; and a failure to recognise that preferences, and the decisions that
result from them, are fundamentally constructive in nature. We conclude by
outlining a decision-aiding approach that has been shown to improve the
quality of decisions about carbon management.
Keywords: decision making; carbon management; climate change; CCS;
policy.
Reference to this paper should be made as follows: Campbell-Árvai, V.,
Bessette, D., Kenney, L. and Árvai, J. (2019) ‘Improving decision making for
carbon management initiatives’, Int. J. Risk Assessment and Management,
Vol. 22, Nos. 3/4, pp.342–358.
Biographical notes: Victoria Campbell-Árvai is an Research Scientist in the
University of Michigan’s School for Environment and Sustainability (SEAS).
Douglas Bessette is an Assistant Professor for Energy Systems in Michigan
State University’s Department of Community Sustainability.
Lisa Kenney is a Smart Mobility Advisor for the Greater Buffalo Niagara
Regional Transportation Council in Buffalo.
Joseph Árvai is the Director of the Erb Institute and the Max McGraw
Professor of Sustainable Enterprise in the University of Michigan’s School for
Environment and Sustainability and Ross School of Business.
1 Introduction
As awareness and concern about the negative social, ethical, environmental, and
economic consequences of climate change increase, so does the frequency with which
researchers and policy makers must explore strategies to mitigate its effects. These range
from reducing sources of carbon (such as the decarbonisation of energy systems and
supply- and demand-side efficiency measures) to geoengineering Earth’s climate (e.g.,
using stratospheric sulfate aerosols).
Another increasingly popular class of mitigation strategies focuses on increasing
sinks of carbon from the atmosphere by enhancing land-based and oceanic carbon uptake.
A number of natural strategies have received serious attention from researchers and
policy-makers including changing agricultural practices to increase carbon retention –
such as no-till agriculture (Lal et al., 2004) – as well as discouraging deforestation, and
incentivising reforestation, via economic or policy levers (Kenney et al., 2015a).
344 V. Campbel
l
-
Á
rvai et al.
At the same time, interest is growing in artificial means of increasing the number and
capacity of carbon sinks. Garnering significant attention in many industrialised countries
is the strategy of separating carbon from fossil fuel emissions and capturing it in the
oceans or underground in geological reservoirs (e.g., in depleted oil fields, coal seams, or
saline aquifers). Several such carbon capture and storage (CCS) (sequestration) projects
are operating or proposed, with many more underground carbon injection initiatives
operational for enhanced oil recovery purposes (GCCSI, 2019).
As with many technological solutions to anthropogenic problems, public concern
about – and opposition to – CCS has been reported. Perhaps the most well-known
example is the opposition to the deployment of CCS in Barendrecht in the Netherlands
(Terwel et al., 2012). But the Dutch are far from alone in their sentiments toward the
technology; negative perceptions of CCS have also emerged in the USA (Krause et al.,
2014), Canada (L’Orange Seigo et al., 2014), France (Ha-Duong et al., 2009), Spain
(Oltra et al., 2010), Switzerland (Wallquist et al., 2009), the United Kingdom (Shackley
et al., 2004), and elsewhere across Europe (Upham and Roberts, 2011).
In order to overcome opposition to CCS, a commonly held assumption among policy
makers is to increase the public’s knowledge of the technology, through risk
communication and environmental education, with a focus on CCS’s risks and benefits
(Ashworth et al., 2012; Hansson and Bryngelsson, 2009; Itaoka et al., 2009; van Alphen
et al., 2007). It is difficult to argue against this viewpoint. When it comes to complex and
uncertain technologies like CCS, our view is also that the quality of decision making (and
stakeholder engagement) will be improved if decision makers know more about the
problems and opportunities they are being asked to consider. However, at the same time,
research in the decision sciences suggests that more knowledge and better information
will not necessarily lead to better decisions about specific CCS initiatives, or endorsing
CCS in general (Arvai, 2014b).
Decision researchers have long demonstrated that in loosely structured situations,
people struggle with a predictable set of difficulties when making complex decisions
about technologies like CCS. One set of difficulties are behavioural or perceptual in
nature; that is, they are related to how information is presented and framed, and how
intuitive or routinised judgmental processes interact with, and often preempt, more
in-depth analysis (Arvai, 2014a). One of the fundamental conclusions from this research
is that decision makers often end up making choices that only partially address the full
range of their concerns.
A second set of difficulties is associated with the decision problems themselves.
Decisions about carbon management are both complex and involve significant
uncertainties. Making these decisions thoughtfully and comprehensively involves
analysing and evaluating various elements of these decisions, rather than ignoring
elements or simplifying the decision in general. The former requires pulling the decision
problems apart into more cognitively manageable components, yet without simplifying
them to the point that the resulting choices ignore their multi-attribute nature (Bessette
et al., 2014; Gregory et al., 2012).
With these two difficulties as a backdrop, this paper reviews five obstacles to
improved decision making about carbon management initiatives. These obstacles stand
out because we encounter them – individually or in combination – in all manner of
decisions, involving all kinds of decision makers. The first of these obstacles is linked to
human behaviour and perceptions: in particular, decision makers’ overreliance on
judgmental shortcuts leading to systematic biases in decision making. The remaining four
I
mproving decision making for carbon management initiatives 345
obstacles deal with the complexity of the problems themselves. These include neglecting
the objectives and related measurement criteria, which will both guide decisions among
competing risk management options; the tendency to look for singular solutions to
complex problems, rather than considering a broad array of options; a lack of explicit
attention devoted to the full range of tradeoffs that should be considered when choosing
among alternatives; and a failure to recognise that preferences, and the decisions that
result from them, are fundamentally constructive in nature. We conclude by outlining a
decision-aiding approach that has been shown to improve the quality of decisions about
carbon management.
2 Decision-making obstacles
2.1 Judgmental shortcuts and systematic biases
A commonly held view in the behavioural sciences is that humans are fundamentally
rational decision makers who will make choices in order to maximise utility or welfare.
To illustrate this point, in decisions that involve risk and uncertainty, rational decision
makers should choose the alternative with the highest net benefit considering all of the
attributes that define all of the alternatives under consideration. Accordingly, evaluating
alternatives so as to assess overall utility would involve a series of rather complex and
time consuming calculations that proponents of rational choice assume decision makers
can make quickly and well (Stanovich, 2010).
An alternative viewpoint is that decision makers – because of task complexity
coupled with limited processing ability – are unable to be fully rational. Instead, decision
makers’ capabilities are ‘bounded’ (Simon, 1972; Simon, 1955) such that they make the
best of their decisions by evaluating a much tighter set of considerations (Gigerenzer and
Selten, 2002). To make these decisions even more efficiently, decision makers have been
observed relying heavily upon a series of intuitive heuristic principles that reduce
complex judgments into much simpler operations (Gilovich et al., 2002; Kahneman et al.,
2011). The advantage of judgmental shortcuts is that they reduce the amount of time, as
well as the level of effort, required to make decisions without compromising well-being,
especially for many routine decisions that demand low levels of effort and accuracy
(Johnson and Payne, 1985; Payne et al., 1993). However, a reliance on judgmental
shortcuts also poses a serious challenge to high quality decision making: As the context
of the choice becomes more complex or unfamiliar, a heavy reliance upon these
heuristics frequently leads to introduction of systematic biases that lead decision makers
away from choices that are closely aligned with their stated priorities.
A commonly applied judgmental heuristic1 in judgments and decisions about
unfamiliar technologies (such as CCS) is known as availability. Here, decision makers
make judgments about the likelihood of certain events and outcomes based on the ease
with which they can be brought to mind, and not based on the actual probability
distribution of the outcomes in question. The bias is introduced when the reasons
underlying the salience of outcomes – e.g., prominent coverage of certain events by the
media – is not well calibrated with their real-world probability distribution (Tversky and
Kahneman, 1973). And because probabilities and outcomes are the key factors in
computations of risk, the availability heuristic often leads decision makers to
mischaracterise that risk.
346 V. Campbel
l
-
Á
rvai et al.
Research suggests that the cancellation of the CCS project proposed by Royal Dutch
Shell for Barendrecht was, in part, a result of strongly negative public reactions to the
project. These reactions, in turn, were heavily influenced by a high-profile public
relations campaign (against the CCS project and its proponents) and negative portrayals
in the local media (Ashworth et al., 2012; Terwel et al., 2012).
A similar situation unfolded in Canada during 2011 following claims of CO2 leakage
from an enhanced oil recovery operation and CCS monitoring site near the town of
Weyburn, Saskatchewan (Boyd et al., 2013). While the facility near Weyburn continued
to operate, other Canadian demonstration projects that had been planned (e.g., a
university-sponsored project near Calgary, Alberta) were casualties of the controversy. In
both the Barendrecht and Weyburn cases, thoroughly reviewed scientific estimates about
the probabilities and risks associated with CO2 leakage could do little to counteract public
risk perceptions based on negative portrayals of CCS.
Another judgmental heuristic likely to be consequential for decisions about carbon
management is based on what psychologists refer to as affect. Affect is defined as an
instinctive emotional state that is invoked when people are confronted with a stimulus.
These include, for example, feelings of joy, dread, and fear. Affect is also a characteristic
that people may instinctively attach to a stimulus, for example qualities such as goodness
or badness (Slovic et al., 2005). The related affect heuristic leads to judgments about
objects, activities, and other stimuli that are shaped by the varying degrees of positive or
negative feelings that they invoke or that people attach to them (Finucane et al., 2000).
The affect heuristic is powerful in that it can readily prevent decision makers from
analysing problems or opportunities in depth. For example, when making choices about
environmental issues, alternatives can prove themselves to be so emotionally charged that
they preempt decision makers from fully evaluating the alternative options based on
whether they are decision makers’ best, long-term interests (Campbell-Arvai et al., 2014;
Wilson and Arvai, 2006).
It should not come as a surprise that the affect heuristic often spills over from
personal decisions into policy decisions (Wilson and Arvai, 2006; 2010). For example,
many decision makers support or reject certain technologies because of the positive or
negative emotions they invoke, and not because of their risks or benefits. A case in point
is atmospheric geoengineering (i.e., solar radiation management via stratospheric sulfate
aerosols), which is rejected by many decision makers on emotional grounds well before
they undertake a more thorough analysis of its feasibility, risks, and benefits (Keith,
2014). As public fears about certain examples of carbon management (e.g., CCS) mount,
it is highly likely that the affect heuristic will play an important role in modulating
judgments and preferences about them.
2.2 Objectives and performance measures
Our research on judgmental heuristics, and the affect heuristic in particular, supports the
dual processing theory of decision making (Epstein, 1994; Zajonc, 1980). The affective
system, often referred to as ‘System 1’, is preconscious and automatic, and attaches
meaning to incoming information based on feelings and emotional connections. The
analytic, rational system – or System 2 – requires decision makers to undertake a much
more conscious appraisal of incoming information. From a neurological perspective
(Montague and Berns, 2002), these systems act in parallel, not in series. As a result, high
I
mproving decision making for carbon management initiatives 347
quality decision making processes must achieve an appropriate balance between the two
systems, such that neither preempts the other (Arvai and Campbell-Arvai, 2013).
In addition to achieving a balance across these two judgmental systems, decision
making processes must also strive to achieve a balance across the range of objectives that
guide decisions. Objectives are the raw materials of decision making, and they reflect
what matters to – and the values of – decision makers (Keeney and Raiffa, 1993). Too
often, however, decision makers rush through the process of identifying the full range of
objectives that guide their choices. The result is that the resulting set of objectives is
incomplete (e.g., decisions about resource development that only consider the objectives
of environmental protection and job creation). The implication of an incomplete set of
objectives is that the overall performance or utility of different courses of action cannot
be accurately determined because other important objectives are omitted from the
analysis (e.g., enhancing community well-being, protecting cultural traditions).
In our own work, we have found strong support from decision makers for taking the
time at the start of a decision making process to identify a comprehensive set of
objectives, which characterise the key values and concerns of decision makers and
stakeholders (Kenney et al., 2015b). These objectives can be visually summarised using
tools such as means-ends networks, objectives hierarchies, value trees and influence
diagrams (Keeney, 1992).
In addition to being comprehensive, objectives should also be understandable; e.g.,
concise and free from ambiguity. It is almost always the case, for example, that decision
makers and stakeholders can benefit from greater resolution around concepts such as
‘sustainability’, which, given the range of knowledge sources in developing community
settings, could have a number of different meanings (Gregory et al., 2012). The
objectives used to guide decisions should also be independent, meaning that each
objective contributes independently to the overall performance of the alternatives under
consideration. Finally, objectives should also be articulated with agreed-upon definitions
and directionality. For example, decision makers often want lower costs, more jobs, and a
higher degree of environmental protection (Keeney and Raiffa, 1993). In thinking about
directionality, decision making processes must also seek clarity on the definition of
actions; e.g., precision in terms of what it means to ‘protect’ or ‘maintain’ certain targets
of concern (Gregory et al., 2012).
Once a set of objectives has been identified, the next consideration is to determine
how each objective will be achieved. In other words, decision makers must identify
indicators that will take the objectives from more abstract statements, to concrete ways of
measuring (or forecasting) the anticipated performance of the alternatives under
consideration. Like objectives, these performance measures should be complete and
concise, and should address the range of consequences associated with a given alternative
without obscuring the outcomes with extraneous data. Likewise, performance measures
should be unambiguous and understandable, with all stakeholders in agreement about the
measures’ meaning, magnitude and direction (Gregory et al., 2001).
Finally, decision making processes must be explicit about the kinds of performance
measures that are required. Performance measures come in three forms: natural,
constructed, and proxy (Keeney and Gregory, 2005). Natural measures tend to be
measured in universally recognised units like costs (e.g., in dollars), and directly measure
the objective in question. Constructed measures must be created if there are no obvious or
universally agreed upon single measures available. Issues such as the quality of life,
perhaps in areas where carbon management is to be deployed, would fall into this
348 V. Campbel
l
-
Á
rvai et al.
category. Here, multiple measures are often combined to produce a single index, and
numbers are used to reflect different levels of a qualitative or descriptive scale. Finally,
proxy attributes are indirect measures of objectives, and are used when neither natural nor
constructed measures are available. An improvement in public health is an oft-cited
objective and the rates of certain illnesses are commonly used as a proxy measure.
2.3 Multiple options
Too often carbon management initiatives are framed as ‘take-it-or-leave-it’ alternatives.
For example, should the geologic storage of CO2 be adopted in a certain area, yes or no?
This approach poses a significant obstacle to high-quality decision making in that
‘decisions’ about contentious issues with single alternatives are akin to ultimatums that
do little to help decision makers grapple with the multi-attribute and multi-solution nature
of complex problems (i.e., how to manage the sharp increase in greenhouse gas
emissions). Rather than focusing on single alternatives, decision making processes about
carbon management should offer a set of complete and comparable (proposed) solutions
to a problem or opportunity (Kenney et al., 2015b).
A second challenge is that too often the development of options focuses too narrowly
on what stakeholders or decision makers perceive to be idealised courses of action. In our
view, the development of alternatives should focus on identifying courses of action that
are substantially different from one another, thereby giving decision makers the
opportunity to learn about how different objectives will be influenced by diverse
alternatives. The bottom line here is that the identification of alternatives should be
viewed as an opportunity to explore a wide range of creative solutions; even those that
may at first seem implausible (Keeney, 1992).
A third challenge manifests itself when carbon management initiatives, which are
required because of the consequences of certain actions, are isolated from the causes of
the emissions. Because carbon management alternatives are always implemented in
conjunction with other activities (e.g., resource development in the case of enhanced oil
recovery, or carbon capture during energy development), our research suggests that
decision makers must begin to look for portfolios of options rather than individual actions
(Bessette et al., 2014).
This means undertaking an objectives-based evaluation of technologies like CCS in
conjunction with – and, importantly, without – the activities that lead to the greenhouse
gas (GHG) emissions in the first place. To this end, he US National Academy of Sciences
(NRC, 2009), amongst others (Socolow et al., 2004), have suggested that decision
making tools with the added capability of developing and analysing combinations of
alternatives – e.g., different energy generation technologies combined with different
kinds and levels of carbon management technologies – be developed so that decision
makers may take a systems approach to achieving GHG emissions reductions. For this
reason, the inclusion of portfolio analysis alongside other decision support tools is
increasingly being viewed as a critical addition to the suite of approaches used for
developing and evaluating carbon management options (Bessette et al., 2014; Choptiany,
2017; Fleishman et al., 2010; Palmgren et al., 2004).
I
mproving decision making for carbon management initiatives 349
2.4 Tradeoffs
Decisions with multiple objectives require that decision makers confront tradeoffs, which
is an explicit exploration of how much of something decision makers are willing to give
up in exchange for getting something else. In the realm of carbon management initiatives
such as CCS, these tradeoffs may be considered across objectives. Or they might be
based on different tolerances for uncertainty and risk; that is, how much loss, uncertainty,
or risk are decision makers willing to accept in terms of one dimension of a decision in
order to achieve gains in another dimension?
Confronting tradeoffs is difficult from both a psychological and practical perspective.
Psychologically, tradeoffs that make decision makers feel as though they must subvert
some morally significant values in favour of others, which invokes a sense of constitutive
incommensurability (Tetlock et al., 2000). Many decision makers react to this conflict by
avoiding tradeoff analysis to the extent that they can. It is common, for example, for
decision makers to consider only a small handful of objectives – sometimes just a single
objective – when making choices; see the discussion of bounded rationality above. It is
important to note that the objectives left out of consideration are not necessarily judged to
be unimportant on their own. Rather, these objectives often are important but have been
discounted because they are difficult or uncomfortable to balance against other attributes
(Arvai et al., 2012).
From a practical perspective, confronting tradeoffs poses challenges because it is hard
to know where and how to begin. When tradeoffs are confronted, it is instinctive to think
of them in general terms; e.g., a willingness to compromise some measure of economic
growth in exchange for some measure of environmental protection. But in decisions
about carbon management, the measures associated with these objectives are much more
specific. For example, does a specific price on stored carbon – e.g., $60 tCO2 stored –
justify going forward with a CCS project as a means of curbing the deleterious effects of
climate change? If not, what about $120 tCO2 stored? Moreover, how does the tradeoff
calculus change over time as additional CCS projects are brought online?
An additional complication for carbon management initiatives is that there are many
more than two objectives in play at any given time. In addition to environmental and
economic objectives, decision makers are asked to consider issues related to human
health, environmental justice, flexibility, and social well-being. These too must be
accounted for using performance measures that quantify achievement under different
alternatives (see above), and must be traded-off when making choices about carbon
management initiatives (Bessette and Arvai, 2018). And, as the number of objectives in
play increases so too does the level of complexity involved in confronting tradeoffs
(Clemen, 2004).
In light of these challenges, decision researchers – ourselves included – have been
experimenting with different kinds of decision-support tools that help decision makers to
explore tradeoffs in a manner that is direct and comprehensive (in that all of the
objectives are considered). For example, we have been combining energy systems models
and interactive computer interfaces to help decision makers develop and forecast the
performance of portfolios. We have also gone beyond visual displays in developing
decision-support tools that help decision makers to more thoughtfully confront on an
objective-by-objective basis, e.g., see Bessette et al. (2014).
350 V. Campbel
l
-
Á
rvai et al.
2.5 Constructed preferences
Taken together, the previous four sections challenge the common assumption held by
many behavioural scientists (and project proponents) that decision makers have a series
of pre-existing preferences about carbon management initiatives that simply need to be
activated through the provision of information about a project’s risks and benefits. This is
an incorrect assumption. Rather than simply reviewing information about a decision
problem and the available options, identifying the appropriate pre-existing preference,
and then making a choice, research shows that decision makers construct their judgments
across a wide range of decision contexts – often from scratch – during elicitation
processes (Lichtenstein and Slovic, 2006; Payne et al., 1992).
Generally, these decision contexts share one or more of three characteristics. First, the
decision context may be novel, with the implication that pre-existing preferences cannot
possibly exist. Second, decision makers are almost always faced with a situation in which
the evaluation of competing alternatives causes conflict between two or more pre-existing
preferences or objectives. In these cases, tradeoffs become necessary, which requires the
construction of new preferences that balance conflicting priorities. Third, decision makers
may be required to translate qualitative expressions of preference into quantitative ones
(or vice versa). Moving from a general judgment that CCS is a desirable strategy to
determining one’s willingness to pay for it (e.g., as an additional levy on utility bills)
requires a constructive process.
Decisions about carbon management initiatives include all three of these elements. In
this context, the vast majority of decision makers are unable to evaluate decision
problems and alternatives by simply drawing on pre-existing preferences. Instead, they
must construct their preferences in response to cues that are available during the decision
making process itself (Arvai et al., 2012; Lachapelle et al., 2014). Some of these cues will
be internal, reflecting deeply held worldviews or ideologies. And some will be external,
in the sense that they are associated with the information that accompanies a decision
problem. For example, these may become apparent in light of local, regional or even
national or international events (e.g., media coverage as in the case of the Barendrecht
and Weyburn examples).
The constructive nature of preferences means that decision support processes must be
designed to help improve the quality of resulting choices. The explicit recognition that
decision makers rely heavily on contextual cues that are available to them as they
construct judgments makes it possible for analysts and facilitators to provide a more
defensible and helpful context – in other words, a structure – for decision making.
Indeed, it is our opinion that proponents of carbon management initiatives – e.g., in
industry and government – are obligated to employ decision-aiding processes that help
decision makers to construct the highest-quality judgments possible in light of the various
constraints they face.
3 Structuring carbon management decisions
Structured decision making (SDM) is a decision support approach that focuses on the
judgmental challenges discussed above – both those that are behavioural and perceptual
in nature, and those that are a product of problem and task complexity. SDM helps to
decompose complex decisions, thereby addressing judgmental challenges common to
I
mproving decision making for carbon management initiatives 351
individual and group deliberation, which are outlined above. In doing so, SDM highlights
the objectives, alternatives, and required tradeoffs in order to provide maximum insight
about the choices that decision makers face (Arvai et al., 2012; Bessette et al., 2014;
Bessette and Arvai, 2018; Kenney et al., 2015b).
SDM relies on the normative benchmarks for high quality decision making that come
from multiattribute utility theory (Keeney and Raiffa, 1993) and decision analysis
(Clemen, 2004) to frame the stakeholder engagement process. In addition, a SDM
approach draws on insights from behavioural decision research to identify common
biases in decision making that must be addressed in order to maximise decision quality
(Kahneman et al., 1982; National Research Council, 2005). SDM also draws on the
guidance of and good practice in risk communication (Arvai and Campbell-Arvai, 2013;
Arvai and Rivers, 2013; Leiss and Larkin, 2018) and deliberation (National Research
Council, 1996) so as to elicit information from stakeholders and technical experts in
credible and defensible ways.
SDM comprises five2 key elements (Gregory et al., 2001; Hammond et al., 1999):
1 defining the decision problem that is to be the focus of analysis
2 identifying stakeholders’ and decision makers’ objectives that will guide the decision
making process, including the performance measures that will be used to gauge
success or failure
3 creating a broad array of creative alternatives that directly address objectives
4 forecasting the consequences associated with alternative courses of action, including
key sources of uncertainty
5 helping decision makers to confront value tradeoffs that arise when selecting among
alternatives.
A range of analytic tools is available for use during an SDM process. These include
decision trees and influence diagrams for characterising decision points and alternatives,
objectives hierarchies for displaying the values and concerns of stakeholders, portfolio
builders for creating alternatives, consequence matrixes for depicting the
objective-by-objective performance of alternatives, and software tools for tradeoff
analysis. When SDM processes are led by analysts and facilitators, they often rely upon
interviews, workshops, surveys, and experiments (e.g., choice and factorial experiments)
to collect necessary information about objectives and preferences (Arvai et al., 2014)
However, SDM can also be practiced by individuals and groups working on their own
(Hammond et al., 1999).
We have studied and applied SDM for carbon management decisions in North
America, as well as abroad. In a pair of applied studies, we worked with colleagues to
develop and test an interactive decision-aiding framework designed to help decision
makers construct internally consistent preferences, while also providing stakeholders (and
decision makers) an opportunity to thoughtfully construct and compare carbon
management portfolios (Bessette et al., 2014; Bessette et al., 2016). To make these
studies as realistic as possible, we worked in a real-world decision context; the focus of
both studies was to inform decisions about energy transitions in communities looking to
replace aging coal-fired generation infrastructure. In order to make decisions about
carbon management relevant, different approaches for reducing GHG emissions (e.g.,
CCS, demand-side management) could be treated separately, or coupled with different
352 V. Campbel
l
-
Á
rvai et al.
energy generation options (i.e., fossil-fuel based power plants, as well as non GHG
emitting generation options, like small modular nuclear reactors).
Both studies involved the development of a software interface, which
1 provided necessary background information to users regarding the development of
coupled carbon management-energy development strategies
2 accounted for users’ values and objectives
3 allowed for the construction of bespoke portfolios bounded by real-world
technological, supply, and demand constraints
4 provided a more rigorous basis for addressing tradeoffs.
This latter element was especially important as it facilitated choices that were in line with
decision makers’ stated priorities. Three analytic approaches were used to evaluate the
framework: participants’ self-reports of satisfaction, the amount and type of knowledge
participants gained, and the overall decision quality as a function of the internal
consistency of participants’ decisions (Bessette et al., 2014; Bessette et al., 2016).
Overall, participants in these studies reported3 very high levels of satisfaction and
comfort with their decisions. An analysis of the data from one of the studies (Bessette et
al., 2014) suggested that these high satisfaction ratings were linked to the fact that the
objectives, and performance measures, used to guide the decision making process were
elicited from the community where the study took place. The data from both studies also
suggested that the ability to both create and evaluate a wide range of energy and carbon
management portfolios was both helpful from the standpoint of ensuring a transparent
process and helping decision makers to untangle what was a rather complex problem.
Specifically, participants who built portfolios and compared them with many others felt
they had been provided with significantly more and yet also just the right amount of
information about energy development and carbon management to make their decisions.
By contrast, participants who did not have an opportunity to build portfolios reported
wanting more information (Bessette et al., 2014; Bessette et al., 2016).
Both studies also showed that the SDM process, with its emphasis on tradeoff
analysis, was also instrumental in leading to internally consistent decisions; according to
many decision researchers, internal consistency (more than self-reported satisfaction) is
the most accurate metric for measuring decision quality (Clemen, 2004; Keeney and
Raiffa, 1993). Recall from Section 2.1 that one of the challenges associated with the use
of judgmental heuristics is that they often lead to choices that do not always reflect
decision makers’ objectives. For this reason, explicit attention to tradeoffs across
alternatives and objectives serves as an important de-biasing tool (Bessette et al., 2014;
Bessette et al., 2016).
A second, related project borrowed insights from the work of Bessette et al. (2014,
2016) in an effort to utilise SDM in guiding the development of a coupled energy-carbon
management strategy in Canada’s Northwest Territories (Kenney et al., 2015b). The
initial focus of this work suggested ways that decision makers in the Northwest
Territories (NWT) might better structure decisions that will require portfolios of solutions
that address decision makers’ and stakeholders’ (e.g., First Nations, public, government,
and industry) objectives across a broad geographic scale.
As with the previous example, this work emphasised educating decision makers about
judgmental biases, the elicitation of objectives and performance measures, creating
I
mproving decision making for carbon management initiatives 353
portfolios of alternatives, and tradeoff analysis. But, unlike the previous example, this
SDM effort is not treating carbon management as a single discreet decision. In most
places, not just the NWT, it will be through interlinked regional and national carbon
management strategies that real progress on curbing GHG emissions can be made.
In terms of results
4, the application of the SDM process in the NWT helped
government decision makers to move beyond a simplified characterisation of carbon
management as something that can be applied in a standardised fashion across the
territory. This, in turn, led to the recognition that not only would portfolios of
technologies be needed at each geographic location; different portfolios of options would
need to be developed for different geographic regions in the territory. The application of
the SDM process also helped decision makers to clarify the range of objectives that guide
energy and carbon management decisions across the territory, as well as the performance
measures that could be used to evaluate portfolios of alternatives. Finally, decision
makers in the territory were provided with tools that would help them to confront
tradeoffs, which we believe will lead to more internally consistent choices moving
forward.
A third study applied SDM in a developing country setting, to a forest-based carbon
management program in Vietnam (Kenney et al., 2015a). Our work utilised a series of
SDM workshops with both national and village level stakeholders in an effort to identify
objectives, along with associated performance measures, and to begin designing
alternatives for the carbon management program. While stakeholders at both levels
generally agreed on the core objectives, they differed in terms of the ways to achieve
these objectives; this highlights the need for locally-specific decision making processes
that effectively capture the range of relevant objectives.
As the SDM process in Vietnam progressed, we received positive feedback from
village level participants who expressed a sense of ownership over the program design
process. Participants at the national level also provided positive feedback about SDM as a
useful participatory method for designing, managing and implementing this program in
Vietnam.
In the end, our research suggests that SDM serves as an effective approach for
involving multiple stakeholders in complex and often contentious issues facing
developing communities. Likewise, SDM brings much-needed methodological precision
to decision contexts that increasingly demand science-based rigour. All of this must
unfold against the backdrop of the manifold challenges that typify developing
communities: poverty, limited infrastructure, low literacy levels, a diversity of cultural
norms and local traditions, poor coordination among decision makers, and – all too
frequently – a lack of democratic and transparent decision making processes. SDM,
therefore, is a way to help decision making processes surrounding carbon management –
in both developing and developed country contexts – recognise and respond to these
challenges.
4 Conclusions
We have focused in this paper on challenges to, and the structure for, decision making
about carbon management. We are not the first to take an interest in these issues. Others
who have studied decision making processes set against the backdrop of human-
environment interactions have identified similar challenges (e.g., dealing with priority
354 V. Campbel
l
-
Á
rvai et al.
setting, and calibrating decisions to these priorities). These, in turn, have led to critiques
of decisions for their perceived failure to meet linked environmental, social and economic
goals (Arvai, 2014a; National Research Council, 2005, 2008).
In spite of these critiques, very little research has focused on developing
science-based guidance regarding how best to address these challenges. Instead, too
many policy makers view decision making as a black box or as art over science. As a
result, we often see very general statements about the ingredients necessary for high
quality decision making: e.g., involving all of the interested parties in decision making
processes, providing timely access to information, ensuring transparency, and the like.
In our view, it will continue to be difficult for policy makers to address complex,
multi-objective decisions – like those about carbon management – without treating
decision making processes in terms that go well beyond general recommendations.
Overgeneralising the decision making process only results in a failure to focus explicitly
on several issues central to higher quality decision making: addressing judgmental biases,
clarifying objectives and associated performance measures, thinking creatively about
alternatives, confronting tradeoffs, and recognising and adapting to the constructive
nature of judgments.
For this reason, one of the most important contributions of a structured decision
making approach is its ability to convey the key elements of a complex problem in terms
that capture its primary impacts – and potential solutions – in ways that can more easily
be understood and addressed. As a result, it becomes easier for decision makers to
understand the expected physical, economic, or social impacts of an action, and
importantly, how they are related. SDM also makes it easier for decision makers to
understand and confront the tradeoffs that must be made when selecting amongst
competing options (Arvai et al., 2001; Bessette et al., 2014; Bessette et al., 2016). This, in
turn, leads to choices about carbon management – as well as other problems at the
human-environment nexus – that do a better job of addressing matters that are of concern
to decision makers (and stakeholders). In our view, this is what higher quality decisions
about carbon management ought to be all about.
Acknowledgements
We wish to thank our many colleagues who have informed our thinking about this topic,
and who assisted us during our research. This research was supported by Carbon
Management Canada, and by the National Science Foundation under award number SES
1728807 to Decision Research and the University of Michigan.
References
Arvai, J. (2014a) ‘An appeal for smarter decisions’, Policy Options, Vol. 35, pp.40–43.
Arvai, J. (2014b) ‘The end of risk communication as we know it’, Journal of Risk Research,
Vol. 17, No. 10, pp.1245–1249.
Arvai, J. and Campbell-Arvai, V. (2013) ‘Risk communication: insights from the decision
sciences’, in Arvai, J. and Rivers, L. (Eds.): Effective Risk Communication: Learning from the
Past, Charting a Course for the Future’, pp. 234–257, Taylor and Francis, London, UK.
Arvai, J. and Rivers, L. (2013) Effective Risk Communication: Learning from the Past, Charting a
Course for the Future, Taylor and Francis, London, UK.
I
mproving decision making for carbon management initiatives 355
Arvai, J., Gregory, R., Bessette, D. and Campbell-Arvai, V. (2012) ‘Decision support for the
development of energy strategies’, Issues in Science and Technology, Vol. 28, No. 4,
pp.43–52.
Arvai, J., Kellon, D., Leon, R., Gregory, R. and Richardson, R. (2014) ‘Structuring international
development decisions: confronting trade-offs between land use and community development
in Costa Rica’, Environment, Systems and Decisions, Vol. 34, No. 2, pp.224–236.
Arvai, J.L., Bridge, G., Dolsak, N., Franzese, R., Koontz, T., Luginbuhl, A., Robbins, P.,
Richards, K., Smith Korfmacher, K., Sohngen, B., Tansey, J. and Thompson, A. (2006)
‘Adaptive management of the global climate problem: bridging the gap between climate
research and climate policy’, Climatic Change, Vol. 78, No. 1, pp.217–225.
Arvai, J.L., Gregory, R. and McDaniels, T. (2001) ‘Testing a structured decision approach:
value-focused thinking for deliberative risk communication’, Risk Analysis, Vol. 21, No. 6,
pp.1065–1076.
Ashworth, P., Bradbury, J., Wade, S., Ynke Feenstra, C.F.J., Greenberg, S., Hund, G. and
Mikunda, T. (2012) ‘What’s in store: lessons from implementing CCS’, International Journal
of Greenhouse Gas Control, Vol. 9, pp.402–409, DOI: 10.1016/j.ijggc.2012.04.012.
Bessette, D., Arvai, J. and Campbell-Arvai, V. (2014) ‘Decision support framework for
developing regional energy strategies’, Environmental Science and Technology, Vol. 48,
No. 3, pp.1401–1408.
Bessette, D., Campbell-Arvai, V. and Arvai, J. (2016) ‘Expanding the reach of participatory risk
management: testing an online decision-aiding framework for informing internally consistent
choices’, Risk Analysis, Vol. 36, No. 5, pp.992–1005.
Bessette, D.L. and Arvai, J.L. (2018) ‘Engaging attribute tradeoffs in clean energy portfolio
development’, Energy Policy, Vol. 115, pp.221–229, DOI: 10.1016/j.enpol.2018.01.021.
Boyd, A.D., Liu, Y., Stephens, J.C., Wilson, E.J., Pollak, M., Peterson, T.R., Einsiedel, E. and
Meadowcroft, J. (2013) ‘Controversy in technology innovation: contrasting media and expert
risk perceptions of the alleged leakage at the Weyburn carbon dioxide storage demonstration
project’, International Journal of Greenhouse Gas Control, Vol. 14, pp.259–269,
DOI: 10.1016/j.ijggc.2013.01.011.
Campbell-Arvai, V., Arvai, J. and Kalof, L. (2014) ‘Motivating sustainable food choices: the role
of nudges, value orientation, and information provision’, Environment and Behavior, Vol. 46,
No. 4, pp.453–475.
Choptiany, J.M. (2017) ‘A holistic multi-criteria decision and risk analysis method for CCS and
GHG mitigation projects’, in: Raydugin, Y. (Ed.): Handbook of Research on Leveraging Risk
and Uncertainties for Effective Project Management, pp.384–405, IGI Global, Hershey, PA.
Clemen, R.T. (2004) Making Hard Decisions: An Introduction to Decision Analysis, PWS-Kent
Publishing Co., Boston, MA.
Epstein, S. (1994) ‘Integration of the cognitive and the psychodynamic unconscious’, American
Psychologist, Vol. 49, No. 8, pp.709–724.
Finucane, M.L., Alhakami, A., Slovic, P. and Johnson, S.M. (2000) ‘The affect heuristic in
judgments of risks and benefits’, Journal of Behavioral Decision Making, Vol. 13, No. 1,
pp.1–17.
Fleishman, L.A., De Bruin, W.B. and Morgan, M.G. (2010) ‘Informed public preferences for
electricity portfolios with CCS and other low-carbon technologies’, Risk Analysis, Vol. 30,
No. 9, pp.1399–1410.
GCCSI (2019) Global CCS Institute [online] https://www.globalccsinstitute.com.
Gigerenzer, G. and Selten, R. (2002) Bounded Rationality: The Adaptive Toolbox, MIT Press,
Boston, MA.
Gigerenzer, G., Hertwig, R. and Pachur, T. (2011) Heuristics, Oxford University Press, Oxford,
UK.
Gilovich, T., Griffin, D. and Kahneman, D. (2002) Heuristics and Biases: The Psychology of
Intuitive Judgment, Cambridge University Press, Cambridge, UK.
356 V. Campbel
l
-
Á
rvai et al.
Gregory, R., Arvai, J.L. and McDaniels, T. (2001) ‘Value-focused thinking for environmental risk
consultations’, Research in Social Problems and Public Policy, Vol. 9, pp.249–275.
Gregory, R., Failing, L., Harstone, M., Long, G., McDaniels, T. and Ohlson, D. (2012) Structured
Decision Making: A Practical Guide to Environmental Management Choices,
Wiley-Blackwell, Chichester, UK.
Gregory, R., Ohlson, D. and Arvai, J.L. (2006) ‘Deconstructing adaptive management: criteria for
applications to environmental management’, Ecological Applications, Vol. 16, No. 6,
pp.2411–2425.
Ha-Duong, M., Nadaï, A. and Campos, A.S. (2009) ‘A survey on the public perception of CCS in
France’, International Journal of Greenhouse Gas Control, Vol. 3, No. 5, pp.633–640 [online]
http://dx.doi.org/10.1016/j.ijggc.2009.05.003.
Hammond, J., Keeney, R.L. and Raiffa, H. (1999) Smart Choices: A Practical Guide to Making
Better Decisions, Harvard Business School Press, Cambridge, MA.
Hansson, A. and Bryngelsson, M. (2009) ‘Expert opinions on carbon dioxide capture and storage –
a framing of uncertainties and possibilities’, Energy Policy, Vol. 37, No. 6, pp.2273–2282,
DOI: 10.1016/j.enpol.2009.02.018.
Itaoka, K., Okuda, Y., Saito, A. and Akai, M. (2009) ‘Influential information and factors for social
acceptance of CCS: the 2nd round survey of public opinion in Japan’, Energy Procedia,
Vol. 1, No. 1, pp.4803–4810, DOI: 10.1016/j.egypro.2009.02.307.
Johnson, E.J. and Payne, J.W. (1985) ‘Effort and accuracy in choice’, Management Science,
Vol. 31, No. 4, pp.395–414.
Kahneman, D., Lovallo, D. and Sibony, O. (2011) ‘Before you make that big decision...’, Harvard
Business Review, June, Vol. 89, pp.50–60.
Kahneman, D., Slovic, P. and Tversky, A. (1982) Judgment Under Uncertainty: Heuristics and
Biases, Cambridge University Press, Cambridge, UK.
Keeney, R. and Gregory, R. (2005) ‘Selecting attributes to measure the achievement of objectives’,
Operations Research, Vol. 53, No. 1, pp.1–11.
Keeney, R.L. (1992) Value Focused Thinking. A Path to Creative Decision Making, Harvard
University Press, Cambridge, MA.
Keeney, R.L. and Raiffa, H. (1993) Decisions with Multiple Objectives: Preferences and Value
Tradeoffs, Cambridge University Press, Cambridge, UK.
Keith, D.W. (2014) A Case for Climate Engineering, MIT Press, Boston, MA.
Kenney, L., Arvai, J., Vardhan, M. and Catacutan, D. (2015a) ‘Bringing stakeholder values into
climate risk management programs: decision aiding for REDD in Vietnam’, Society and
Natural Resources, Vol. 28, No. 3, pp.261–279.
Kenney, L., Bessette, D. and Arvai, J. (2015b) ‘Improving decisions about energy strategies in
developing communities: a case study from Canada’s north’, Journal of Environmental
Planning and Management, Vol. 58, No. 5, pp.855–873.
Krause, R.M., Carley, S.R., Warren, D.C., Rupp, J.A. and Graham, J.D. (2014) ‘“Not in (or under)
my backyard”: geographic proximity and public acceptance of carbon capture and storage
facilities’, Risk Analysis, Vol. 34, No. 3, pp.529–540, DOI: 10.1111/risa.12119.
L’Orange Seigo, S., Arvai, J., Dohle, S. and Siegrist, M. (2014) ‘Predictors of risk and benefit
perception of carbon capture and storage (CCS) in regions with different stages of
deployment’, International Journal of Greenhouse Gas Control, Vol. 25, pp.23–32 [online]
http://dx.doi.org/10.1016/j.ijggc.2014.03.007.
Lachapelle, E., Montpetit, É. and Gauvin, J-P. (2014) ‘Public perceptions of expert credibility on
policy issues: the role of expert framing and political worldviews’, Policy Studies Journal,
Vol. 42, No. 4, pp.674–697.
Lal, R., Griffin, M., Apt, J., Lave, L. and Morgan, M.G. (2004) ‘Managing soil carbon’, Science,
Vol. 304, No. 5669, p.393.
I
mproving decision making for carbon management initiatives 357
Leiss, W. and Larkin, P. (2018) ‘Risk communication and public engagement in CCS Projects: the
foundations of public acceptability’, International Journal of Risk Assessment and
Management, in this issue.
Lichtenstein, S. and Slovic, P. (2006) The Construction of Preference, Cambridge University Press,
Cambridge, UK.
Montague, P.R. and Berns, G.S. (2002) ‘Neural economics and the biological substrates of
valuation’, Neuron, Vol. 36, pp.265–284.
National Research Council (1996) Understanding Risk: Informing Decisions in a Democratic
Society, National Academy Press, Washington, DC.
National Research Council (2005) Decision Making for the Environment: Social and Behavioral
Science Research Priorities, The National Academies Press, Washington, DC.
National Research Council (2008) Public Participation in Environmental Assessment and Decision
Making, National Academies Press, Washington, DC.
National Research Council (2009) America’s Energy Future: Technology and Transformation,
National Academies Press, Washington, DC.
Oltra, C., Sala, R., Solà, R., Di Masso, M. and Rowe, G. (2010) ‘Lay perceptions of carbon capture
and storage technology’, International Journal of Greenhouse Gas Control, Vol. 4, No. 4,
pp.698–706 [online] http://dx.doi.org/10.1016/j.ijggc.2010.02.001.
Palmgren, C.R., Morgan, M.G., Bruine de Bruin, W.n. and Keith, D.W. (2004) ‘Initial public
perceptions of deep geological and oceanic disposal of carbon dioxide’, Environmental
Science and Technology, Vol. 38, No. 24, pp.6441–6450, DOI: 10.1021/es040400c.
Payne, J.W., Bettman, J.R. and Johnson, E.J. (1992) ‘Behavioral decision research: a constructive
processing perspective’, Annual Review of Psychology, Vol. 43, pp.87–132.
Payne, J.W., Bettman, J.R. and Johnson, E.J. (1993) The Adaptive Decision Maker, Cambridge
University Press, Cambridge, MA.
Shackley, S., McLachlan, C. and Gough, C. (2004) ‘The public perception of carbon dioxide
capture and storage in the UK: results from focus groups and a survey’, Climate Policy,
Vol. 4, No. 4, pp.377–398, DOI: 10.1080/14693062.2004.9685532.
Simon, H. (1972) ‘Theories of bounded rationality’, in McGuire, C.B. and Radner, R. (Eds.):
Decisions and Organization, North Holland Pulishing Company, Amsterdam.
Simon, H.A. (1955) ‘A behavioral model of rational choice’, Quarterly Journal of Economics,
Vol. 69, pp.99–118.
Slovic, P., Peters, E., Finucane, M.L. and MacGregor, D.G. (2005) ‘Affect, risk, and decision
making’, Health Psychology, Vol. 23, No. 4, pp.s35–s40.
Socolow, R., Hotinski, R., Greenblatt, J.B. and Pacala, S. (2004) ‘Solving the climate problem:
technologies available to curb CO2 emissions’, Environment, Vol. 46, pp.8–19.
Stanovich, K.E. (2010) Decision Making and Rationality in the Modern World, Oxford University
Press, Oxford, UK.
Terwel, B.W., ter Mors, E. and Daamen, D.D.L. (2012) ‘It’s not only about safety: beliefs and
attitudes of 811 local residents regarding a CCS project in Barendrecht’, International Journal
of Greenhouse Gas Control, Vol. 9, pp.41–51, DOI: 10.1016/j.ijggc.2012.02.017.
Tetlock, P.E., Kristel, O.V., Elson, S.B., Green, M.C. and Lerner, J.S. (2000) ‘The psychology of
the unthinkable: taboo tradeoffs, forbidden base rates, and heretical counterfactuals’, Journal
of Personality and Social Psychology, Vol. 78, No. 5, pp.853–870.
Tversky, A. and Kahneman, D. (1973) ‘Availability: a heuristic for judging frequency and
probability’, Cognitive Psychology, Vol. 5, pp.207–232.
Upham, P. and Roberts, T. (2011) ‘Public perceptions of CCS: emergent themes in pan-European
focus groups and implications for communications’, International Journal of Greenhouse Gas
Control, Vol. 5, No. 5, pp.1359–1367 [online] http://dx.doi.org/10.1016/j.ijggc.2011.06.005.
358 V. Campbel
l
-
Á
rvai et al.
van Alphen, K., van Voorst tot Voorst, Q., Hekkert, M.P. and Smits, R.E.H.M. (2007) ‘Societal
acceptance of carbon capture and storage technologies’, Energy Policy, Vol. 35, No. 8,
pp.4368–4380 [online] http://dx.doi.org/10.1016/j.enpol.2007.03.006.
Wallquist, L., Visschers, V.H.M. and Siegrist, M. (2009) ‘Lay concepts on CCS deployment in
Switzerland based on qualitative interviews’, International Journal of Greenhouse Gas
Control, Vol. 3, pp.652–657.
Wilson, R.S. and Arvai, J.L. (2006) ‘When less is more: how affect influences preferences when
comparing low and high-risk options’, Journal of Risk Research, Vol. 9, No. 2, pp.165–178.
Wilson, R.S. and Arvai, J.L. (2010) ‘Why less is more: exploring affect-based value neglect’,
Journal of Risk Research, Vol. 13, pp.399–409.
Zajonc, R.B. (1980) ‘Feeling and thinking: preferences need no inferences’, American
Psychologist, Vol. 35,zzz pp.151–175.
Notes
1 A treatment of the wide range of judgmental heuristics, and their implications for decision
making, is well beyond the scope of this article; for a more thorough review, we direct readers
to other sources, see Gigerenzer et al. (2011), Gilovich et al. (2002) and Kahneman et al.
(1982).
2 A sixth step in SDM, which is described in some publications involves the implementation of
selected alternatives, see Arvai et al. (2012) and then the adaptive management of them as part
of a detailed monitoring process, see Arvai et al. (2006) and Gregory et al. (2006).
3 For a detailed accounting of all results from this study, see Bessette et al. (2014).
4 Unlike the work of Bessette et al. (2014), this work was of a purely applied nature; hence,
empirical measures of decision-makers’ satisfaction internal consistency were not available;
see Kenney et al. (2015b).