External costs of nuclear energy: a pedigree analysis of assumptions

Introduction

In Belgium, as in most industrialized countries, debate about the place of nuclear power production in a sustainable energy mix has resulted in enduring and intractable conflict between actors holding antagonistic positions. External cost calculations can be seen as part of attempts to provide an ‘objective’ input in order to rationalize decision making regarding sustainable energy. In this respect, the most thorough body of work took place within the European Commission (EC) supported ExternE project.
The controversial character of external cost and data and their limited use in policy making processes can be linked to a number of interrelated factors: the complexity of the policy issue and the plural positions, values, and stakes that are into play; the proposed role for external cost data in decision making; the valueladen character of the approach itself.  To explore if, and how, external cost data can contribute in a more effective way to policy making, the Belgian nuclear research centre (SCK-CEN) organized a workshop, in collaboration with the European Commission’s Joint Research Centre and Utrecht University. Crucial assumptions and choices made in the calculation chain were discussed and qualified in a structured way, using the concept of ’‘pedigree of knowledge’. Participants in the workshop included policy makers, stakeholders, and experts from various disciplines. The focus of the discussion was on assumptions related to the scenario used for evaluating the impacts of a severe nuclear accident, to the estimation of the related health impacts, and to their economic and monetary valuation. Special attention was paid to the value-laden character of these assumptions. 

Method

The workshop followed the logic that the relevant uncertainties to consider in relation to external costs, as expressed through the existence of different scientific opinions and/or divergent interpretations and use of results in the socio-political debate, are those who find their source in different value-laden choices, assumptions, and framings entering into the (long) chain of calculations leading to an estimation of externalities. 
A procedure was developed to select key assumptions in a process involving all workshop participants. During the workshop, these assumptions were discussed and qualified using a scheme based on the concept of pedigree of knowledge. As a pilot, it was decided to focus specifically on the externalities of a potential large-scale nuclear accident in a Belgian NPP.

After going through the ExternE research reports and gathering information on points of disagreement and controversy, preparatory interviews with ExternE experts and stakeholders were held, allowing to list in a structured way the main steps in the calculation chain, as well as the crucial assumptions in each of these steps. In this way, a list of 30 assumptions was arrived at, and, through an internet survey, held weeks before the workshop and involving all workshop participants, a final list of six assumptions was selected.  As a result, the focus of the discussions was on assumptions related to the scenario used for evaluating the impacts of a severe nuclear accident, to the estimation of the related health impacts, and to their economic and monetary valuation. Special attention was paid to the value-laden character of these assumptions.

 
To evaluate and qualify these assumptions a scheme was used, developed by Kloprogge et al. (in press). Based on a literature study, and on the concept of pedigree, they discerned a set of criteria to discuss and qualify the potential valueladenness and the influence of assumptions on the end results of an assessment. Value laden here is not to be exclusively understood as politically and/or ethically controversial. It refers to the fact that making an assumption involves going through a choice process.

The criteria used to discuss the assumptions were:
• Influence of situational limitations: The degree to which the choice for the assumption can be influenced by situational limitations, such as limited availability of data, money, time, software, tools, hardware, and human resources.
• Plausibility: The degree, mostly based on an (intuitive) assessment, through which the approximation created by the assumption is in accordance with ‘reality’.
• Choice space: The degree to which alternatives were available to choose from when making the assumption
• Agreement among peers: The degree to which the choice of peers is likely to coincide with the analyst’s choice.
• Agreement among stakeholders: The degree to which the choice of stakeholders is likely to coincide with the analyst’s choice’
• Sensitivity to the view and interests of the analyst: The degree to which the choice for the assumption may be influenced, consciously or unconsciously, by the view and interests of the analyst making the assumption.
• Influence on results: In order to be able to pinpoint important value-laden assumptions in a calculation chain it is not only important to assess the potential value-ladenness of the assumptions, but also to analyse the influence on outcomes of interest of the assessment.

The qualitative discussion of each assumption was closed by giving a score for the assumption for each of the criteria, as indicated by the pedigree matrix. The scoring was organized as an intersubjective process of negotiation that enables one to summarize the main points of discussion for each criterion, to explain why different participants suggested different scores, and to clarify any ambiguity in the descriptions of the criteria. The pedigree process allowed one to qualify the robustness of the assumptions, as assessed by the participants, as well as to discern options for possibly making alternative assumptions to improve the external cost calculation or make them more meaningful.

Results

Overall the scores given were low, reflecting the stakeholder and expert panel’s scepticism about the assumptions made: generally these were considered not very plausible, subject to disagreement, and to a large extent inspired by contextual factors.
The only real exception to this was the assumption that there exists a linear correlation between exposure to ionizing radiation and health effects, even for very small radiation doses. This was the least contested assumption of the workshop. Three experts agreed that the so-called ‘linear no-threshold hypothesis’ (LNT) constitutes the best scientific basis to regulate the risks of ionizing radiation and that the LNT assumption could be qualified as a precautionary approach to managing radiation risks. This statement was qualified to some extent as a result of the discussion with stakeholders. There was a suggestion, however, that perhaps different dose-effect curves should be used for different fractions of the population.

Clear suggestions for improvements in the external cost calculations were made in relation to the assumptions regarding the NPP location that was used as basis for the calculation, and regarding the neglect of other than atmospheric release routes for the dispersion of radionuclides in case of an accident.
The main criticism of the first assumption was that, because of the specific location of the Belgian NPPs—near major cities with important industrial activities—the results obtained by using a hypothetical location in the middle of Western Europe would (seriously) underestimate the externalities of a potential accident for the Belgian context. Therefore, the critics among the stakeholder group argued that a study of the potential consequences of a severe accident (a ‘Probabilistic Safety Assessment (PSA)—Level 3’) for the specific case of a Belgian NPP would likely bring much more insights into the nature and extent of the resulting externalities. The ExternE experts present at the workshop agreed in principle to this objection.

In relation to the second assumption, a PSA expert present at the workshop explained that contamination of ground and river water by radionuclides in the case of a severe nuclear reactor accident cannot be entirely excluded. Such contamination could be caused by a failure of the NPP’s pressure vessel and a possible melting of the reactor core through the bottom of the reactor building, resulting in steam explosions. This accident sequence is considered to be highly improbable, but would nevertheless result in a long-term ecosystem pollution, which is not included as an externality in the ExternE approach (only impacts on human health count). This view was not really challenged in the discussion.
A second category of assumptions was formed by those that were heavily criticized and for which, only to a certain extent, proposals were made for how external cost calculations could be improved to overcome this criticism. The assumption that the used cost indicators (cost of countermeasures, direct economic damage, short- and longterm health impacts) are representative for the total costs of a potential large-scale nuclear accident in a Belgian NPP...has to be mentioned here.

Mainly, representatives of environmental Non-governmental organizations (NGOs) pointed out that an entire catalogue of economic impacts were not included in the ExternE methodology: direct and indirect costs of lost production in industries adjacent to the NPP, forward ripple effects in the entire European economy (for example, caused by an evacuation of the Antwerp harbour), costs of ‘stigmatization’ of a region contaminated by nuclear fallout, economic impacts on the nuclear sector worldwide (for example, costs of cancelling new nuclear programmes, enhanced safety measures in existing plants, etc.), and so on. The discussion on this assumption led to most of the participants seeing a continued great potential for public contestation of it.

Some assumptions were clearly seen as problematic although it was not clear how to overcome the related problems through other ways of calculation. This category of assumptions included the neglect of all health impacts, other than the radiological ones, caused by exposure to ionizing radiation when assessing the consequences of a large-scale nuclear accident in a Belgian NPP. This was criticized by a radiation protection expert on the basis of the experience after the Chernobyl reactor accident in 1986. Follow-up studies show a significant increase in the population suffering from post-traumatic stress symptoms, anxiety, estrangement, dislocation, etc. However, according to this expert it is difficult (if not impossible) to relate these psychological impacts unequivocally to a nuclear accident as such or rather to the risk management interventions of (Soviet) authorities. Therefore, he concluded that non-radiological (psychosomatic) health impacts of a potential reactor accident could prove to be a major— albeit hardly quantifiable—factor. The overall average pedigree score attributed by the workshop participants to this assumption shows that this expert qualification was largely shared.

Also belonging to this category of assumptions is the acceptability of not reporting a ‘risk-aversion factor’ for accidents of the ‘low probability/high consequences’ type. The ExternE experts present at the workshop admitted that within the ExternE network, a better conceptualization of the ‘risk aversion factor’ for potential severe nuclear accidents was not considered to be a priority issue. Within the ExternE network the prevailing opinion was that since risk aversion is a highly contextual factor that varies enormously from individual to individual, no robust theoretical framework could be advanced to determine the exact extent of the externality resulting from risk aversive attitudes. Participants in the workshop found this to be one of the weakest links in the calculation chain since it is not up to researchers to implicitly decide that individuals or segments of society should not have different degrees of aversion for different categories of risk.

Conclusion

Based on an analysis of the underlying reasons for the limited role of external cost studies in the Belgian debate and decision making on sustainable energy options, and of the nuclear option in particular, it can be argued that traditional patterns of one-way communication in the science–policy interface should be at least complemented, if not abandoned.
The main finding of the workshop is that the underlying fundamental critique of the ExternE approach can only be dealt with adequately through a re-negotiation of the place and role of external cost data in the sustainable energy debate. The assumed ‘ideal’ relation between external cost results and policy measures has to be openly discussed and debated, allowing to surface the implicit meanings and value-laden assumptions behind the uncritical presentation of such direct link. Possibilities for such re-negotiation can be created through direct interaction between involved stakeholders, relevant experts, and policy makers in a setting that allows experimenting with the socio-institutional roles ‘normally’ taken up in the science–policy–society interface. In this respect, the workshop showed the potential of the approach of pedigree assessment, to assess assumptions made in external cost studies.

By using pedigree assesment the following results can be achieved:
• qualifying assumptions when they are communicated to the policy sphere
• evaluating how assumptions relate to different perspectives and frameworks, as held by involved actors
• generating suggestions for improvements in the calculation chain to deal with disagreements and divergence over assumptions as well as suggestions for communicating this type of policy relevant science.
• conveying a more adequate image of policy-relevant science.

Documentation of the case

E. Laes, G. Meskens and J.P. van der Sluijs (2011). On the contribution of external cost calculations to energy system governance: The case of a potential large-scale nuclear accident. Energy Policy, 39 (9) 5664-5673. doi:10.1016/j.enpol.2011.04.016

M. Craye, E. Laes, J. van der Sluijs (2009). Re-negotiating the Role of External Cost Calculations in the Belgian Nuclear and Sustainable Energy Debate. In: A. Pereira Guimaraes and S. Funtowicz. Science for Policy, Oxfor University Press, pp 272-290.

References

P. Kloprogge, J.P. van der Sluijs and A.C. Petersen (in press). A method for the analysis of assumptions in model-based environmental assessments Environmental Modelling & Software
Doi: 10.1016/j.envsoft.2009.06.009

European Commission (EC) (1995), ExternE: Externalities of Energy, Vol. 1–6, Project report EUR 16520–16525, Luxembourg: Office for the
Official Publications of the European Communities. [Volume 5: Nuclear]

European Commission (EC) (1999), ExternE: Externalities of Energy, Vol. 7–10, Project report EUR 19083–19086, Luxembourg: Office for the Official Publications of the European Communities. http://www.externe.info/