Framing risks and uncertainties 17
are not clearly distinct from each other and inevitably overlap, for example in
social and economic impacts. Thus, flexibility and at the same time clarity in
the specific case- based application are paramount.
Methods for assessing risks and uncertainties
All narratives presented in this book draw on quantitative modelling, tightly
interlinked with stakeholder consultations, to develop the transition pathways at
their heart. The general equilibrium and optimisation models used for this purpose
provide outputs as point estimates rather than probabilities, creating a false level
of precision sometimes known as deterministic uncertainty (Nikas, Doukas, and
Papandreou, 2018). Models suffering from this issue may be unsuitable if one
wants to produce strategies that are less vulnerable to a large range of plausible
outcomes. Moreover, for these models it is more difficult to consider implementation than consequential risk. It was thus of utmost importance to involve stakeholders not only for the specification of pathways but also for the identification
and assessment of associated risks (Doukas et al., 2018). For this purpose, we
designed stakeholder engagement processes, all of which built on the same set of
fundamental data collection formats, including: desktop research for the identification and analysis of policy documents; open and semi- structured expert- and
stakeholder interviews (face- to-face, e- mail, telephone); and different focus groups
and workshop formats. At the stakeholder identification and selection stage, we
made sure that different, even opposing perspectives were considered in the analysis, so that different views could be heard and compared. This made the stakeholder consultation an as- good-as- possible representation of the actual fields of
tension in our case study contexts. With stakeholders, we discussed risks and
uncertainties qualitatively in interview and workshop settings, and with the
option of a standardised assessment using Likert scales. This facilitated the characterisation of risks along different dimensions such as likelihood, severity of impact,
timing of impact, mitigation capacity, and general level of concern.
Any of these social- empirical methods comes with certain limitations or
potential biases, which can be reduced but not eliminated. For the best understanding and interpretation of our results, readers should be aware of the most
important biases that potentially affect our research. These are:
(1) the selection of stakeholders: although the aim in each case was to select a
comprehensive set of stakeholders using a stakeholder matrix approach, it is
often not possible to include all relevant voices; and
(2) the capacity of stakeholders to assess risks, as stakeholders may or may not
be experts with respect to the policies in questions.
Therefore, this is not a formal expert elicitation process (Morgan, 2014) aiming
at quantified probabilities for very specific events, but an informed assessment of
perceived potential barriers and negative outcomes. This is particularly the case
for those risks and uncertainties that depend on individual and collective
are not clearly distinct from each other and inevitably overlap, for example in
social and economic impacts. Thus, flexibility and at the same time clarity in
the specific case- based application are paramount.
Methods for assessing risks and uncertainties
All narratives presented in this book draw on quantitative modelling, tightly
interlinked with stakeholder consultations, to develop the transition pathways at
their heart. The general equilibrium and optimisation models used for this purpose
provide outputs as point estimates rather than probabilities, creating a false level
of precision sometimes known as deterministic uncertainty (Nikas, Doukas, and
Papandreou, 2018). Models suffering from this issue may be unsuitable if one
wants to produce strategies that are less vulnerable to a large range of plausible
outcomes. Moreover, for these models it is more difficult to consider implementation than consequential risk. It was thus of utmost importance to involve stakeholders not only for the specification of pathways but also for the identification
and assessment of associated risks (Doukas et al., 2018). For this purpose, we
designed stakeholder engagement processes, all of which built on the same set of
fundamental data collection formats, including: desktop research for the identification and analysis of policy documents; open and semi- structured expert- and
stakeholder interviews (face- to-face, e- mail, telephone); and different focus groups
and workshop formats. At the stakeholder identification and selection stage, we
made sure that different, even opposing perspectives were considered in the analysis, so that different views could be heard and compared. This made the stakeholder consultation an as- good-as- possible representation of the actual fields of
tension in our case study contexts. With stakeholders, we discussed risks and
uncertainties qualitatively in interview and workshop settings, and with the
option of a standardised assessment using Likert scales. This facilitated the characterisation of risks along different dimensions such as likelihood, severity of impact,
timing of impact, mitigation capacity, and general level of concern.
Any of these social- empirical methods comes with certain limitations or
potential biases, which can be reduced but not eliminated. For the best understanding and interpretation of our results, readers should be aware of the most
important biases that potentially affect our research. These are:
(1) the selection of stakeholders: although the aim in each case was to select a
comprehensive set of stakeholders using a stakeholder matrix approach, it is
often not possible to include all relevant voices; and
(2) the capacity of stakeholders to assess risks, as stakeholders may or may not
be experts with respect to the policies in questions.
Therefore, this is not a formal expert elicitation process (Morgan, 2014) aiming
at quantified probabilities for very specific events, but an informed assessment of
perceived potential barriers and negative outcomes. This is particularly the case
for those risks and uncertainties that depend on individual and collective