Comparative socio-cultural analysis of risk perception of Carbon Capture and Storage in the European Union

Energy Research & Social Science 21 (2016) 114–122

Contents lists available at ScienceDirect

Energy Research & Social Science journal homepage: www.elsevier.com/locate/erss

Original research article

Comparative socio-cultural analysis of risk perception of Carbon Capture and Storage in the European Union Farid Karimi ∗ , Arho Toikka, Janne I. Hukkinen Environmental Policy Research Group, University of Helsinki, PO Box 16, 00014 Helsinki, Finland

a r t i c l e

i n f o

Article history: Received 14 August 2015 Received in revised form 27 June 2016 Accepted 29 June 2016 Keywords: CCS National culture Risk perception The general public

a b s t r a c t The transition to a sustainable energy regime is not just an engineering question, but a social and cultural issue as well. In this paper, we consider one contested technology still in development, Carbon Capture and Storage (CCS), from a socio-cultural perspective. CCS is widely deemed to be a necessary bridging technology to a low-carbon economy, but the technology needs to pass considerable hurdles before widespread use. The importance of cultural issues in CCS deployment has been acknowledged, but research on the large-scale cultural patterns is lacking. To fill this knowledge gap, we combine aggregated individual level measurements of technology opinions with indicators that characterize national cultures. We use survey data from a Eurobarometer together with prior cross-cultural data to show that nationspecific cultural issues can be used as a macro-level approximation of public reactions to CCS technology. Public reactions incorporate cultural factors such as the degree of separation between groups, strength of institutions over space, time and social roles, and society’s tolerance for uncertainty and ambiguity. On the basis of the analysis, we provide a richer frame for analysts wishing to understand why and how societies and societal actors challenge and contest technologies and energy regimes. © 2016 Elsevier Ltd. All rights reserved.

1. Introduction Carbon Capture and Storage (CCS) has been proposed as a promising technology to mitigate climate change and facilitate a leap towards sustainability. CCS means the removal of carbon from the atmosphere and storage in carbon sinks. Since the early 2000s, CCS has become the centre of attention in many countries. Given the deceptively abundant amounts of fossil fuel, CCS is considered a medium-run option for reducing CO2 and dealing with climate change [28,17]. The technology is a transition measure to sustainable energy, because from the policymakers’ viewpoint, especially those from fossil-fuel possessing countries, it is an easy way to sustainable development in comparison to other options [28,17,32]. In Europe, the Netherlands and UK are prominent examples of countries where CCS has been pushed by governments and companies alike. CCS has raised controversy as well. Opponents such as Spreng et al. [64] argue that it is a “double-edged sword” and leads to technological lock-in that hinders the development of renewable

∗ Corresponding author at: Environmental Policy Research Group, University of Helsinki, Helsinki, Finland. E-mail addresses: farid.karimi@helsinki.fi (F. Karimi), arho.toikka@helsinki.fi (A. Toikka), janne.i.hukkinen@helsinki.fi (J.I. Hukkinen). http://dx.doi.org/10.1016/j.erss.2016.06.024 2214-6296/© 2016 Elsevier Ltd. All rights reserved.

energies. Social opposition has slowed down CCS development. For example, the Shell Carbon Capture and Storage project in Barendrecht in the Netherlands, which was planned to store 10 million tons of CO2 over a period of 25 years starting in 2011, was cancelled as a result of opposition by the local community [36]. The implementation of the EU CCS Directive encountered opposition in Germany where states were against the proposed on-shore demonstration storage sites. Technologies are culturally embedded, which cannot be overlooked when considering the trajectory of a technology from an innovation to a major component of the energy regime. While plenty of research on the technical issues related to CCS exists, many authors have called for further research on its sociocultural aspects. Vercellia and Lombardi [76,4840] claim that ‘if social culture does not develop in parallel with technology, when CCS will reach the commercial stage it might be felt as an imposition and thus refused’. Most socio-cultural research has focused on local contexts and case studies with relatively few large-scale international comparative studies. Notable exceptions are Reiner et al. [51], the Accsept project [2], Oltra et al. [45], Itaoka et al. [30] and the Eurobarometer survey on CCS [13], the data set we use in this analysis. The Eurobarometer data set enables us to view the socio-cultural aspects of CCS development from a novel perspective, namely,

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

national culture. Although some cultural theorists are sceptical of aggregating cultural features on the basis of nation states, environmental and technology policies tend to be aligned with national characteristics. From a policy perspective, then, it would make sense to uncover the variation of risk perception and reactions to CCS technology across countries. The Eurobarometer data does precisely this. It measures individual knowledge, individual opinion and background socio-demographic variables, and enables us to make comparative analyses of how people perceive the risk and react to technological developments, and how these perceptions and reactions differ between countries. In this paper we demonstrate that macro-scale cultural issues are one explanatory factor in how individuals (on average) perceive the risks and benefits of new technologies. We show that dimensions of national culture have consequences for CCS development beyond the level of local communities immediately affected by projects. Projects are embedded in larger cultural contexts, and countries with different social settings need to be accounted for when developing an understanding of the relationship between risk perception and the acceptability of CCS technology. We understand public acceptance as a complex mixture of individual perceptions of risks and benefits that can be analysed quantitatively. We use measures of national culture in combination with survey data on acceptance – in the broad sense of individual reactions to hypothetical or real projects and technologies – to demonstrate this argument. We understand culture as ‘the collective programming of the mind that distinguishes the members of one group or category of people from another’ [19,9]. National culture is not uniform, and it is not perfectly measurable, but commonly available indices of how complex issues such as uncertainty, power, and individualism are manifested in different countries and are available as an approximation. If one wants to compare the differences between a small number of countries based on some other explanatory factor, like the status of CCS or knowledge of CCS, these cultural differences are a potential confounding factor in the country comparison. We argue that deeper and more long-standing issues shape the immediate and observed risk perceptions and other explanatory factors such as trust, one of them being the macro-culture we analyse here. We address two questions: How does national culture influence the risk and benefit perceptions of CCS? To what extent is public reaction to the implementation of CCS predictable in a crosscultural comparative framework? To answer the questions, we employ survey data from the Eurobarometer on Public Awareness and Acceptance of CO2 capture and storage, a large scale survey conducted in twelve European countries. We operationalize the cultural factors with Hofstede and Minkov’s [25] cultural dimensions theory, a six-dimensional summary of national culture used in an extensive assortment of fields including cross-cultural psychology, cross-cultural communication, political science and risk research [48,6,82,80,4]. We correlate the average opinions in each country with its cultural dimensions as defined and measured by Hofstede and Minkov to understand the rationale behind reactions to CCS implementation in different countries.

2. Background 2.1. CCS and earlier research on socio-cultural issues In this section we provide an overview of recent research on the socio-cultural aspects of CCS. In particular, we aim to specify the following key concepts: risk perception, public acceptability, cultural factors, and national culture. We conclude the discussion with a hypothesis of the relationships between the concepts that will guide our empirical research.

115

Not enough is known of the relationship between the pattern of laypeople risk perception and macro-level cultural factors affecting it. For instance, Ashworth et al. [3] show that despite sharing the same concerns and having been exposed to the same information about CCS, citizens from different countries tend to adopt different perceptions of the technology. Similarly, Pietzner et al. [50] conducted a representative survey in six EU countries to assess public awareness and perceptions of CCS and concluded that public perceptions and awareness of CCS vary considerably in different countries. L’Orange Seigo et al. [38] review of 42 research articles on public perception of CCS recommends further research on the role of social context and values in the deployment of CCS. Cross-cultural factors have been overlooked in CCS research, and in public perception studies of CCS the emphasis has mainly been on supplying information, communication, knowledge dissemination and trust (e.g. [71,27,29,67,68,8,30,12]). These are crucial elements but they do not reveal the full picture, as the unexplained country differences noticed in the studies reviewed above show. Despite extensive research on the social and political issues of CCS [43,41,72–75], and many authors referring to the role of cultural features in the perception and/or deployment of the technology, few studies incorporate explicit cultural variables. For instance, Bradbury [7] discusses public perception of CCS by analysing six CCS projects in the US. Her broad conclusion is that for the deployment of CCS, ‘differences in social and cultural framework’ have to be accommodated. Similarly, Oltra et al. [45] studied public reaction to CO2 storage sites in five EU countries and note that culture is one of the factors influencing risk perception, but do not elucidate further. Wallquist et al. [79] recommend further research on the role of value-based trust in various contexts and on other beliefs that explain variance in risk perceptions of CCS. The Social Licence to Operate framework, which has recently been applied to CCS and extended to regional and national contexts, measures the level of ongoing approval and societal acceptance by local communities [16]. Zhang et al. [86] show that various aspects of approval and factors affecting approval vary by context, and suggest that further research is needed on how culture affects these issues. The meaning and dimensionality of laypeople responses to CCS vary in social research. Some studies have worked with the contested notion of unidimensional public acceptance [78,29]. The notion of single accept/refuse dimension has been criticized for misrepresenting complex assemblages such as the ways in which public opinion arises in social processes [65] and misrepresenting the negotiation between multiple societal goals and the use of a variety of technologies to reach those goals, and the distribution of harms and benefits related to such alternatives [85]. Issues such as trust towards actors and governance structure the public’s response to projects [27]. Public acceptance is, then, a property of a particular decision, not a property of a technology. For example, Wallquist et al. [78] measure acceptance on a continuous scale from ‘not at all acceptable’ to ‘totally acceptable’. While this measure is sensible in comparisons between related technologies, as used in Wallquist et al. [78], it does not predict actual response to a proposal, as the respondents are not asked to consider issues such as trust, funding, ownership, risk aversion and so forth that would be included in a specific decision. Others have simply queried people if they would favour the use of CCS or particular parts or varieties of it in their local community [15] or country [59]. Other studies have opened up the concept of acceptability with two- and multidimensional models. Such latent variable approaches have been used with survey data. Itaoka et al. [29] find two latent dimensions, general acceptance and geological storage acceptance. Tokushige et al. [71] surveyed Japanese university students in a confirmatory factor analysis based on a six-factor acceptance model: public acceptance, risk perception, benefit per-

116

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

ception, trust, views on human interference with nature in CCS, as well as global warming in general. Acceptance was strongly influenced by the perceived benefits, while the estimated likelihood of various technical risks affected acceptance slightly negatively. Huijts et al. [27] measure perceived risks and benefits, attitudes, emotions and trust in actors to form a big picture of acceptance. To untangle the complexities of the concept of public acceptability, we adopt the position of Wallquist et al. [77], who argue that analysing risk and benefit perception instead of acceptability generates more differentiated insights. Surveying laypeople by directly asking whether they accept or support a technology is unlikely to give meaningful results. In an effort to strike a balance between respecting the nuances of different social constructions of risk and operationalizing such nuances for practical decision making, we rely on quantitative analysis and define the factors of public acceptability of CCS empirically based on the Eurobarometer data. Hence, we propose a two-dimensional representation of laypeople reactions, to be discussed in Section 3. Some researchers are sceptical about social acceptance studies whose methodology is based on surveys of social attitudes toward science and technology [46,10]. Laypeople who barely have a notion of what a technology entails are said to hold only pseudo-opinions; fickle impressions easily changed by the supply of information and biased by the supplier of this information [40]. We agree that opinions held are not the results of weighing pros and cons and deeply thinking about the issue. Even ‘the public’ is generated as a response to a particular issue [5]: in politics, an established process exists for going from independent voters to final policy, but in surveys, this process has to be imagined as if the average opinion or a vocal minority were meaningful gauges of what would happen after the nation engaged in actual deliberation of the issue. Survey responses are a proxy measure for such actual deliberation, and may be changed easily by further information. However, as the research reviewed above has found, how these changes happen varies by culture, and it is important to know how. On the basis of earlier research we hypothesize that the public acceptability of CCS is a function of several factors relating to risk and benefit perception. These factors are in turn influenced by several cultural factors, which vary fairly robustly from one nation to another. Thus, national cultural features are reasonable candidates for explaining variation in risk and benefit perception and therefore also the public acceptability of CCS. We now turn to the question of operationalizing the concept of national culture.

2.2. Cultural dimensions Research into perception, acceptance and adaptation of technologies in different countries needs to account for different natural and technical settings as well as differences in the setting in which people live, frame their worldviews and form their opinions. Multiple theories exist on how cultures differ and multiple ways to operationalize these differences into quantitative measurements. However, any model of national culture is necessarily a compromise between representativeness and usefulness. Hofstede’s six-dimensional model is a good compromise in that the measured concepts match well with what would be expected to be important specifically in the context of technology (Table 1). Hofstede’s dimensions are designed to highlight differences between national cultures and can contribute to explaining reasonable variances across countries. In addition, some of Hofstede’s dimensions fit aptly to study issues related to risk perception and contingencies. For example, Uncertainty Avoidance, defined by Hofstede as the extent to which individuals feel uncomfortable about unknown, uncertain and unclear situations or occasions [25], or Masculinity vs. Femininity, are dimensions directly relevant to the study of risk perception and understanding of climate change related technologies, as will be shown soon. Hofstede developed his framework empirically. The indicators are based on factor analysis of IBM employees’ survey data [39]. Subsequent analyses have modified the initially four-dimensional space of culture. Afterwards, he added a fifth and sixth dimension based on Bond’s Chinese Value Survey study and Michael Minkov’s research [25]. In the bipolar dimensions of Table 1, such as Individualism vs. Collectivism or Masculinity vs. Femininity, the increasing measurements refer to the first concept. For instance, in Masculinity vs. Femininity, a country with a higher score is a masculine society and a country with a lower score is a feminine society. The quantification and operationalization of culture, especially national culture, are very controversial subjects. The dimensionalization of culture is criticized, mainly from the anthropological point of view, which argues that it is not possible to quantify the shared beliefs, attitudes and values of a society based on surveys and statistical data as in the work of Hofstede and Minkov [18,84,47]. Translation error, cultural differences in response style and representativeness of survey sample are claimed to be the shortcomings of cross-cultural dimensions, along with stereotyping [18,47]. In addition, some researchers criticize Hofstede’s

Table 1 Hofstede’s cross-cultural dimensions adopted from Maleki and de Jong [39]. Value axis

Explanation

Individualism vs. Collectivism (IDV)

Individualism stands for a society in which the ties between individuals are loose. It means a person is expected to look after himself/herself and his/her immediate family only. Collectivism stands for a society in which people from birth onwards are integrated into strong, cohesive in-groups, which continue to protect them throughout their lifetime The extent to which the less powerful members of society expect and accept that power is distributed unequally The extent to which members of society feel uncomfortable with uncertain, unknown, ambiguous, or unstructured situations. The key factor here is how a society deals with the fact that the future can never be known. Masculinity represents a preference in society for achievement, competition, heroism, assertiveness and material reward for success. Femininity stands for a preference for cooperation, consensus, modesty, caring for the weak and quality of life. Long-term Orientation stands for a society that fosters virtues and is oriented towards future rewards, in particular perseverance and thrift. Short-term orientation refers to a society that fosters virtues related to the past and present, in particular respect for tradition, preservation of ‘face’, and fulfilling social obligations. Indulgence represents a society that allows relatively free gratification of basic and natural human drives related to enjoying life and having fun. Restraint stands for a society that suppresses gratification of needs and regulates it by means of strict social norms.

Power Distance (PDI) Uncertainty Avoidance (UAI) Masculinity vs. Femininity (MAS)

Long vs. Short term Orientation (LTO)

Indulgence vs. Restraint (IVR)

Note: Despite its prima facie impression, the dimension Masculinity vs. Femininity (MAS) should not be misconstrued as a cultural index pertinent to gender-related characteristics. To avoid the confusion, other researchers have suggested alternative terms, e.g., Mastery vs. Harmony [39]. However, to adhere to the common terminology of the discipline, this dimension is referred to as Masculinity vs. Femininity in this paper.

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

methodology in defining some dimensions, such as Individualism vs. Collectivism [66,18]. Hofstede and Minkov have responded to these criticisms [24,23,22,20,21,19,25,44]. One of the main aims of dimensionalization of national culture is to highlight and distinguish the similarities and differences between countries [44]. The assumption is not that IBM surveys are representative of each society, but that using a similar sampling procedure in each country provides reliable estimates of the differences between countries [19]—the difference between IBM staff in two countries is probably similar to the difference between the unobserved true representative scores of the nations. The dimensions have been validated using completely different sources. These include not just surveys but also non-survey qualitative data such as McClelland’s [42] data of content analysis of children books [19]. The dimension of Individualism vs. Collectivism has been extracted from various samples including students [9], company employees [63], teachers [58] and nationally representative samples [81]. The accusation of stereotyping is claimed to be unfounded, because scores of national culture are not stereotypes for individuals but rather for national societies [25,44]. We also considered using Mary Douglas’s cultural theory as an analytical framework. However, studies based on Douglasian cultural theory focus on generic cultural features that have little to do with national characteristics and that relate instead to the degree to which an individual belongs to a social group and is circumscribed by social rules [11,31,60,83,70]. Sjöberg [55,57] even asserts that Douglasian cultural theory is not successful in explaining risk perception and that “a new and quite different approach is needed”. 3. Method and data We use the Eurobarometer data in combination with Hofstede’s national culture indicators to demonstrate the relationship between culture and public reaction to CCS. We find the latent structure of individual reactions to CCS and compare the mean national scores of individual reactions to the cultural indicators. The Eurobarometer CCS survey was conducted to understand awareness and acceptance levels [13]. It asked respondents to rate their knowledge of climate change and CCS and their views on CCS. The survey was conducted as part of the Eurobarometer survey programme in 2011 and included 13,901 respondents spread over 12 European countries where CCS projects were started or planned (Table 2). In each country, a multi-stage random sampling procedure was followed. Geographical areas were sampled in a demographically balanced manner, and interviewees were sampled randomly within these areas. The samples of around one thousand respondents in each country were balanced with poststratification weights to account for non-response, so that each national sample corresponds to the population of the country [13].

Since we had to rely on secondary data, the selection of countries was fixed. However, we believe that the countries represent a reasonable mix of cultures, as all five European country clusters from a meta-analysis of cultural typologies are represented [52]. The countries do differ in their adoption of CCS, but the proportion of informed citizens was fairly similar in all countries except the Netherlands, where the issue is much more debated [13]. The Netherlands did not appear substantially different in an outlier analysis, so we are fairly confident that the observed cultural differences are not due to country differences in CCS adoption. The respondents were queried on their opinions over a variety of CCS issues, but not directly on acceptance or whether they accept the use of the technology either in general or for specific projects. Rather, they were asked if they were concerned about the risks, whether they would benefit personally or regionally, or if CCS would help combat climate change. In principle, a person could present almost any combination of opinions and would still find a CCS project at the local coal plant perfectly reasonable. For example, even if you believe CCS does not help combat climate change and CCS will not lower energy prices, it is logical to have no open negative reaction to someone else building a CCS plant. It could be argued that the survey only measured correlates of acceptance and not any concept that could be called acceptance. Still, the questions do represent some disposition towards the technology, and there is likely to be some underlying structure in the responses. We analysed all CCS questions that could be understood to represent a quantifiable attitude towards CCS (i.e. we left out those questions where respondents were given the opportunity to describe why they were worried, and so on). Principal component analysis (PCA) is a data reduction method that transforms correlated variables into a smaller set of components that shows which variables reflect which dimensions and how well those variables do measure those dimensions. We chose to use these Likert items and binary variables as if they were continuous in our PCA. Technically, these variables violate the distributional assumptions, but PCA has been shown to be fairly robust to this particular deviation [34]. If anything, estimates on for example variance explained are biased downwards. We discuss this limitation further in Section 5. We pruned the selection of questions a bit further for statistical and research interest reasons. One question, the statement “Fossil fuels will still be used after the year 2050 for electricity production in the EU” was not correlated with any of the other variables, and thus did not contribute to the PCA and was dropped. Using the whole set of questions, a three-component solution appeared appropriate. One of these was interpretable as an attitude towards the role of public authorities in CCS management. We chose to remove these questions on the grounds that they are somewhat two-pronged: they measure whether the respondent wants CCS to be regulated and whether they want it to be regulated by the

Table 2 Status of CCS in 12 EU countries.

Bulgaria Czech Republic Finland France Germany Greece Italy Netherlands Poland Romania Spain UK

Sample size

Reason for inclusion in survey

1001 1004 1001 1035 1622 1000 1027 1012 1000 1053 1004 1322

European funded projecta Heavily dependent on coal Onshore storage not feasible, no projects ongoing Several project ongoing Pilot trials ongoinga Heavily dependent on coal Legislative framework available, pilot-type project ongoing Projects ongoing Projects ongoinga Large-scale projects planned Projects in early stages Projects ongoing

Source: Eurobarometer [13]. a Cancelled or put on hold since the survey.

117

118

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

Table 3 The results of principal component analysis. Component loading

Effective in fighting climate change Will help combat climate change Lowers and stabilizes energy prices Would benefit from local use Would be worried if a local storage site was built A safety risk in the future Variance explained

Risk Perception

0,81 0,80 0,74 0,72 −0,21 −0,21 42,7%

−0,26 −0,22 −0,02 −0,25 0,85 0,85 21,2%

current authorities (i.e. saying you are against the harmonization of CCS methodologies by EU could mean you are against the harmonization or that you are against the EU doing it). Public trust is important in risk perception, but we felt that these questions could not properly untangle the different issues and chose to focus on a narrower, six-question set with two principal components. We were mostly interested in the attitudes presented and not in the absence of opinion, so we removed any ‘do not know’ responses. However, given the high number of those, we validated our analysis by testing how it would change if all those agnostic would be against instead, i.e. recoded don’t know as the anti-CCS option on the survey, and the structure of the results did not change considerably, though extreme attitudes obviously were more common. The results of the final PCA are presented in Table 3. The component loadings are the correlations between the raw variables and the components produced, while the communalities are the proportion of the total variance in the each original variable explained by the two factors. A convention is to consider loadings above .6 high [33], so all our variables fit the model well. We interpret the two generated components as Risk Perception (RP) and Benefit Perception (BP). RP measures how much worry or concern respondents see in CCS use. BP measures a perceived benefit, whether from a personal perspective or as a climate change mitigation tool. We recognize that there are two contrary approaches to the measurement of the risk perception. One approach, which we adopt here, adheres to the psychometric paradigm and considers the role of affection and emotions on risk perception [14,1]. An example of this approach is the “affect heuristic” hypothesis [61,62,37], according to which emotional reactions, including worry, are integrated into other calculable factors that are embedded in risk perception and should hence be considered in the evaluation of risk perception. In contrast, Sjöberg [54–56] proposes a dichotomy in subjective risk perceptions between rational assessment of risk and emotional reactions in terms of expressing worry. Our goal was to come up with nationally representative scores for each country. Next, we saved the principal component score variables for each individual, and took means by country to compare them with the prevalent cultural dimensions in those countries the Eurobarometer asked the CCS questions. The component scores are standardized with mean zero and standard deviation one. In the field of cultural research, researchers have long been cognizant of the differences between cultural phenomena that are simply the sum of individual level features and emergent phenomena that only exist or make sense when viewed from a collective perspective [49]. Empirically, this distinction can mean that a factor analysis on individual data may find a completely different structure compared to one run on a higher-level data, even if it was acquired by taking means of individual measurements and using them to represent the country. To justify our aggregation procedure, we compared our solution to one on the country

1

Principal component extraction, orthogonal rotation.

Communality

Benefit Perception

0,65 0,65 0,57 0,53 0,72 0,73 1

level (using means on each individual variable in a PCA) and individually in each country separately and found few changes to the component structure—the magnitude and direction of the loadings remained the same in all approaches. Thus, we feel confident that the country means of the individual level analysis are a reasonable representation. The scores of Hofstede’s dimensions of national culture (Table 1) are only meaningful in a comparative framework, which we provide for selected European nations (Fig. 1). As we discussed in Section 2.2, the scores of cultural dimensions (Fig. 1) were computed by taking country averages of IBM survey statements and factor analysing them. For example, short-term orientation means agreeing with statements such as ‘effort should produce quick results,’ ‘respect to tradition’ and ‘investment in mutual funds’ [25]. In contrast, long term oriented nations believe in ‘thrift,’ saving money for future and ‘investment in real estate’ [25]. Countries with low uncertainty avoidance score high on statements such as ‘uncertainty is a normal feature of life, and each day is accepted as it comes’ [25] while nations with higher uncertainty avoidance go along with statements like ‘what is different is dangerous’. In addition, higher uncertainty avoidance means that ‘people have more worries about health and money’. The most important feature of a society with high uncertainty avoidance is that ‘there is hesitancy toward new products and technologies’ [25]. Long-term orientation is reflected in preference for perseverance and thrift, while short-term orientation means respect for tradition, importance social obligations and stability.

4. Results We correlated the average factor scores in each country derived from factor analysis (Table 3) with the six cultural dimensions of Hofstede (N = 12). Benefit Perception is high in societies with large power distance, high collectivism, high uncertainty avoidance, short-term orientation and disposition for restraint (Table 4; Fig. 2). Moreover, it is likely that people accept risks in countries with small power distance, higher individualism, femininity, low uncertainty avoidance and high indulgence (Table 4; Fig. 3).

5. Discussion Our analysis makes two contributions. First, to our knowledge this is the first analysis of large-scale cross-cultural patterns of CCS reactions. The small number of countries and the cross-sectional nature of the analysis, however, do not allow for far-reaching specific conclusions about the effects of a particular cultural factor. Nonetheless, the analysis identifies an important omitted variable bias in existing literature, especially in cross-sectional and comparative research. Without explicitly controlling for the setting that survey responses are embedded in, relationships between knowledge and risk/benefit perceptions can be spurious. Second, our study strengthens the interpretation that risk and benefit percep-

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

119

Fig. 1. Cross-cultural scores for 12 European countries included in the Eurobarometer.

Fig. 2. Benefit Perception and the six cultural dimensions.

Table 4 Correlations between country means on the CCS attitude indices and cultural factors.

Power Distance (PDI) Individualism vs. Collectivism (IDV) Masculinity vs. Femininity (MAS) Uncertainty Avoidance (UAI) Long vs. Short-term Orientation (LTO) Indulgence vs. Restraint (IVR) a

Benefit Perception

Risk Perception

0,69a −0,56 0,17 0,33 −0,36 −0,62a

−0,29 0,32 −0,48 −0,60a −0,22 0,31

Correlation is significant at the 0.05 level (2-tailed).

tions should be measured separately, and these phenomena are embedded in cultural context each in their own ways. Our results point to important cultural phenomena that CCS developers have to take into account. Risks are not just accepted

and benefits are not just calculated against them, but rather the cultural underpinnings of the members of a society have important effects. According to Malone et al. [40], when making decisions people account for authority, fit with worldview, salience, and emo-

120

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

Fig. 3. Risk perception and the six cultural dimensions.

tional appeal. The first two of these are social concepts not easy to grasp on an individual level—neither as measurable concepts in research nor as practical concepts in stakeholder involvement. National culture can help to serve as a proxy for them, even at the individual level. The national level correlations do not reveal the social mechanisms of benefit and risk perceptions or the processes by which they arise, but they do show interesting patterns that could explain the events around CCS projects. For example, in long term oriented nations that believe in ‘thrift,’ saving money for future and ‘investment in real estate’ [25], risk aversion seems likely. This puts the Netherlands in an intriguingly contradictory position. On one hand, one would expect the Netherlands as a thrifty nation to expresses a higher level of risk perception regarding CCS. On the other hand, since the Netherlands is an indulgent nation with low distance to power, one would expect it to display more positive attitude than opposition to CCS. Yet it is home to one of the most famous CCS controversies, the Barendrecht case. We propose that the resistance can be explained with the Not In My Back Yard (NIMBY) syndrome: popular resistance to CO2 storage near residences occurs, because the project might have adverse effects on the real estate market. Regardless of national level support for the technology, on-shore storage near residential areas in a thrifty nation concerned about real estate values is the Achilles’ heel of CCS implementation. This is congruent with the results of the Eurobarometer [13] survey, according to which 23% of Dutch respondents had concerns about ‘possible drop in local property prices’—one of the highest rates in the survey. In addition, this is in accord with Krause et al. [35] who hypothesized that owning a dwelling remarkably influences CCS risk perception whereas it has less impact on the national level acceptance of CCS. In other words, initial supporters of CCS deployment change to opposition when a proposed facility moves near their communities. Similarly, a survey among 112 Dutch residents residing in a gas field region that is also a potential CO2 storage site revealed fairly positive attitudes about CCS in general with slightly negative to neutral opinion about CO2 storage in the immediate locale [26].

For another example, Indulgence vs. Restraint (IVR) and Risk Perception indicator correlate with each other. Indulgent nations seem to have a lower level of risk perception toward CCS, which can be attributed to having a ‘positive attitude’ and ‘high optimism’ [25] typical of these societies. For instance, when judged on the basis of cultural characteristics, the Finnish society (IVR = 57) and the British (IVR = 69) might well be among the most supportive countries of CCS implementation while the Bulgarian (IVR = 16) and the Italian (IVR = 30) ones might oppose it. Again, the results of the Eurobarometer [13] survey lend support to this conclusion. When asked to react to a deep underground storage site for CO2 being located within 5 km of their home, 76% of Bulgarian and 62% of Italian respondents said they are ‘worried’. Finland by 50% and the United Kingdom by 52% had the lowest percentages among the other countries. The results of this study also lend support to the Terwel [67], Terwel et al. [68,69] and Dowd et al. [12] hypotheses regarding role of trust in risk perceptions and public acceptability of CCS. In lower power distant and indulgent nations (e.g. Finland and UK) where public trust in government and other stakeholders exists, active opposition from the general public is less likely. This is also in accord with the findings of Shackley et al. [53] which through the observation of a tentative agreement of their interviewees with the inclusion of CCS in the government’s policy target they concluded that a moderate support of CCS exists in the UK. Our analysis has two weaknesses, one relating to sample size and the other to measurement. Given the small sample size of our study, future research should include more countries. As to the second weakness, cross-cultural surveys work well when attitudes, values, and opinions are easily translated and/or the measurements validated in each country. This is rarely the case for contested issues such as novel technologies. Thus, differences in the results of survey comparisons of CCS can arise as much from acceptance definitions associated with national cultures as from the general acceptance believed to be measured. This is also a limitation for our methodological approach: we use survey questions with low-resolution scales that were not directly developed to be measures of the latent components we found. For proper predictive comparisons, we believe the two dimensions found in the Eurobarometer data

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

serve as a good starting point, and risk and benefit measurements should be developed and validated. 6. Conclusion Our first research question was: How does national culture influence the risk and benefit perceptions concerning CCS. We have specified how cultural orientations and their characteristics shape the perception of CCS technology and influence the reactions of the general public. Hierarchical nations that have high uncertainty avoidance are less likely to accept CCS technology. In contrast, feminine nations characterized by social harmony tend to be more welcoming to a technology that aims to increase the long-term quality of life, while masculine societies characterized by social mastery tend to a growing economy which results in deterioration of environmental quality in most cases. Indulgent nations seem more optimistic about CCS. Our second question was: To what extent is public reaction to the implementation of CCS predictable in a cross-cultural comparative framework. The short answer is, it is difficult to conclude that cultural factors alone can predict public acceptability of a CCS project. Local contingencies that may prove fatal to a project are easily overlooked when applying coarse cultural indicators such as those introduced here. However, the analytical framework we have introduced does provide a planning tool for policy makers, authorities and energy companies with which to operationalize the overlooked issue of culture when dealing with the risks and uncertainties of a new energy technology such as CCS. The unique strength of our framework originates in the fact that national specifics weigh considerably in the planning and resource allocation by international organizations. The European Commission, for example, might conceivably benefit from our framework when considering its budget allocation to the member states to implement CCS projects. Our analysis is a first, exploratory one with coarse, nation-level data to demonstrate the principle. The next step in the analysis is to apply a multilevel regression model, with individual level variables such as knowledge of CCS and climate change, and general social variables combined with national level cultural dimensions. When doing a comparative analysis across a number of countries, in which case in-depth qualitative research on particular details of local and national levels is not feasible, the national culture approach can be easily combined with other related frameworks, such the Social License to Operate [86]. While it is interesting to use a variety of variables related to communities’ reactions to new technologies to compare different countries, such comparisons are often limited by the small number of countries and the lack of longitudinal data. Adding cultural factors to these models can help control potential sources of bias. Acknowledgments We thank Ammar Maleki for his comments on cross-cultural analysis. The work was supported by the Academy of Finland project no. 140830 (RICCS) and Finnish Cultural Foundation Professor Pool Grant 2014. References [1] A.S. Alhakami, P. Slovic, A psychological study of the inverse relationship between perceived risk and perceived benefit, Risk Anal. 14 (6) (1994) 1085–1096. [2] J. Anderson, et al., 2007. The ACCSEPT project: Summary of the Main Findings and Key Recommendations. Final report from ACCSEPT. http://www.accsept. org/outputs/wp 5 2dec 2007 final.pdf (accessed 22.08.13). [3] P. Ashwortha, et al., Public preferences to CCS: how does it change across countries? Energy Procedia 37 (2013) 7410–7418.

121

[4] C. Basta, Siting technological risks: cultural approaches and cross-cultural ethics, J. Risk Res. 14 (7) (2011) 799–817. [5] W.L. Bennett, Constructing publics and their opinions, Political Commun. 10 (1993) 101–120. [6] R.N. Bontempo, W.P. Bottom, E.U. Weber, Cross-cultural differences in risk perception: a model-based approach, Risk Anal. 17 (4) (1997) 479–488. [7] J. Bradbury, Public understanding of and engagement with CCS, in: N. Markusson, S. Shackley, B. Evar (Eds.), The Social Dynamics of Carbon Capture and Storage, Routledge, London and New York, 2012, pp. 74–90. [8] S. Brunsting, P. Upham, E. Dütschke, M. De Best Waldhober, C. Oltra, J. Desbarats, H. Riesch, D. Reiner, Communicating CCS: applying communications theory to public perceptions of carbon capture and storage, Int. J. Greenh. Gas Control 5 (6) (2011) 1651–1662. [9] Chinese Culture Connections, Chinese values and the search for culture-free dimensions of culture, J. Cross-Cult. Psychol. 18 (2) (1987) 143–164. [10] M. De Best-Waldhober, D. Daamen, Public Perceptions and Preferences Regarding Large Scale Implementation of Six CO2 Capture and Storage Technologies, Faculty of Social Sciences, Leiden University, 2006. [11] M. Douglas, A. Wildavsky, Risk and Culture, University of California Press, Berkeley, 1982. [12] A.M. Dowd, K. Itaoka, P. Ashworth, A. Saito, M. de Best-Waldhober, Investigating the link between knowledge and perception of CO2 and CCS: an international study, Int. J. Greenh. Gas Control 28 (2014) 79–87. [13] Eurobarometer, Public Awareness and Acceptance of CO2 Capture and Storage, European Commission, Belgium, 2011. [14] B. Fischhoff, P. Slovic, S. Lichtenstein, S. Reid, B. Coombs, How safe is safe enough? A psychometric study of attitudes towards technological risks and benefits, Policy Sci. 9 (1978) 127–152. [15] M. Ha-Duong, A. Nadaï, A.S. Campos, A Survey on the Public Perception of CCS in France, Int. J. Greenh. Gas Control 3 (5) (2009) 633–640. [16] N. Hall, J. Lacey, S. Carr-Cornish, A.M. Dowd, Social licence to operate: understanding how a concept has been translated into practice in energy industries, J. Clean. Prod. 86 (2015) 301–310. [17] Anders Hansson, Mårten Bryngelsson, Expert opinions on carbon dioxide capture and storage—a framing of uncertainties and possibilities, Energy Policy 37 (6) (2009) 2273–2282. [18] S.J. Heine, D.R. Lehman, K. Peng, J. Greenholtz, What’s wrong with cross-cultural comparisons of subjective Likert scales?: the reference-group effect, J. Pers. Social Psychol. 82 (6) (2002) 903–918. [19] G. Hofstede, Culture’s Consequences: Comparing Values, Behaviors, Institutions and Organizations Across Nations, 2nd ed., Sage, Thousand Oaks, CA, 2001. [20] G. Hofstede, Dimensions do not exist: a reply to Brendan McSweeney, Hum. Relat. 55 (11) (2002) 1355–1360. [21] G. Hofstede, The pitfalls of cross-national survey research: a reply to the article by Spector et al. on the psychometric properties of the Hofstede Values Survey Module 1994, Appl. Psychol. Int. Rev. 51 (1) (2002) 170–173. [22] G. Hofstede, What is culture? A reply to Baskerville, Account. Org. Soc. 28 (2003) 811–813. [23] G. Hofstede, Who is the fairest of them all? Galit Ailon’s Mirror, Acad. Manag. Rev. 34 (3) (2009) 570–571. [24] G. Hofstede, The Globe debate: back to relevance, J. Int. Bus. Stud. 41 (8) (2010) 1339–1346. [25] G. Hofstede, G.J. Hofstede, M. Minkov, Cultures and Organizations, 3rd ed., McGraw-Hill, New York, 2010. [26] N. Huijts, Public perception of carbon dioxide storage: the role of trust and affect in attitude formation, in: Master Thesis, Department of Technology Management, Eindhoven University of Technology, 2003. [27] N.M. Huijts, J.M. Cees, A.L. Meijnders, Social acceptance of carbon dioxide storage, Energy Policy 35 (2007) 2780–2789. [28] IPCC, Intergovernmental Panel on Climate Change Special Report on Carbon Capture and Storage, Cambridge University Press, New York, 2005. [29] K. Itaoka, Y. Okudaa, A. Saitoa, M. Akaib, Influential information and factors for social acceptance of CCS: the 2nd round survey of public opinion in Japan, Energy Procedia 1 (2009) 4803–4810. [30] K. Itaoka, A.M. Dowd, A. Saito, M. Paukovic, M. de Best-Waldhober, P. Ashworth, Relating individual perceptions of carbon dioxide to perceptions of CCS: an international comparative study, Energy Procedia 37 (2013) 7436–7443. [31] D.M. Kahan, Nanotechnology and society: the evolution of risk perceptions, Nat. Nanotechnol. 4 (11) (2009) 705–706. [32] F. Karimi, A. Goulas, M. Barzmehri, M. Anggana Putri, CCS potential in Norway—exploring the role of flagship projects: the Mongstad and Kårstø case studies, Int. J. Sustain. Water Environ. Syst. 4 (1) (2012) 23–34. [33] P. Kline, An Easy Guide to Factor Analysis, Routledge, 2014. [34] S. Kolenikov, G. Angeles, Socioeconomic status measurement with discrete proxy variables: is principal component analysis a reliable answer? Rev. Income Wealth 55 (1) (2009) 128–165. [35] R.M. Krause, S.R. Carley, D.C. Warren, J.A. Rupp, J.D. Graham, Not in (or Under) My Backyard: geographic proximity and public acceptance of carbon capture and storage facilities, Risk Anal. 34 (3) (2014) 529–540. [36] L. Limousin, 2010. CCS Communication: Lessons Learnt from Barendrecht. http://www.bellona.org (accessed 20.11.13). [37] G.F. Loewenstein, E.U. Weber, C.K. Hsee, N. Welch, Risk as feelings, Psychol. Bull. 127 (2001) 267–286.

122

F. Karimi et al. / Energy Research & Social Science 21 (2016) 114–122

[38] S. L’Orange Seigo, S. Dohle, M. Siegrist, Public perception of carbon capture and storage (CCS): a review, Renew. Sustain. Energy Rev. 38 (2014) 848–863. [39] A. Maleki, M. de Jong, A proposal for clustering the dimensions of National Culture, Cross Cult. Res. 48 (2) (2014) 107–143. [40] E.L. Malone, J.J. Dooley, J.A. Bradbury, Moving from misinformation derived from public attitude surveys on carbon dioxide capture and storage towards realistic stakeholder involvement, Int. J. Greenh. Gas Control 4 (2) (2010) 419–425. [41] N. Markusson, S. Shackley, B. Evar, The Social Dynamics of Carbon Capture and Storage: Understanding CCS Representations, Governance and Innovation, 1st ed., Routledge, Oxon, 2012. [42] D.C. McClelland, The Achieving Society, Reinhold, Princeton: Van Nostrand, 1961. [43] J. Meadowcroft, O. Langhelle, Caching the Carbon: The Politics and Policy of Carbon Capture and Storage, 1st ed., Edward Elgar Publishing Limited, Cheltenham, 2009. [44] Minkov Minkov, Cross-Cultural Analysis: The Science and Art of Comparing the World’s Modern Societies and Their Cultures, 1st ed., Sage, Thousand Oaks, 2013. [45] C. Oltra, P. Upham, H. Riesch, A. Boso, S. Brunsting, E. Dutschke, A. Lis, Public responses to CO2 storage sites: lessons from five European cases, Energy Environ. 23 (2) (2012) 227–248. [46] R. Pardo, F. Calvo, Attitudes toward science among the European Public: a methodological analysis, Publ. Underst. Sci. 11 (2002) 155–195. [47] K. Peng, R.E. Nisbett, N.Y.C. Wong, Validity problems comparing values across cultures and possible solutions, Psychol. Methods 2 (4) (1997) 329–344. [48] L.S. Pheng, S. Yuquan, An exploratory study of Hofstede’s cross-cultural dimensions in construction projects, Manag. Decis. 40 (2002) 7–16. [49] M.F. Peterson, S.L. Castro, Measurement metrics at aggregate levels of analysis: implications for organization culture research and the GLOBE project, Leadersh. Q. 17 (5) (2006) 506–521. [50] K. Pietzner, D. Schumann, S.D. Tvedt, H.Y. Torvatn, R. Næss, D.M. Reiner, et al., Public awareness and perceptions of carbon dioxide capture and storage (CCS): insights from surveys administered to representative samples in six European countries, Energy Procedia 4 (2011) 6300–6306. [51] D. Reiner, T. Curry, M. de Figueiredo, H. Herzog, S. Ansolabehere, K. Itaoka, M. Akai, F. Johnsson, M. Odenberger, An international comparison of public attitudes towards carbon capture and storage technologies, in: 8th International Conference on Greenhouse Gas Control Technologies (GHGT-8), Trondheim, Norway, 2006. [52] S. Ronen, O. Shenkar, Clustering countries on attitudinal dimensions: a review and synthesis, Acad. Manag. Rev. 10 (3) (1985) 435–454. [53] S. Shackley, C. McLachlan, C. Gough, The public perception of carbon dioxide capture and storage in the UK: results from focus groups and a survey, Clim. Policy 4 (4) (2004) 377–398. [54] L. Sjöberg, World views, political attitudes and risk perception, Risk Health Saf. Environ. 9 (1998) 137–152. [55] L. Sjöberg, Worry and risk perception, Risk Anal. 18 (1998) 85–93. [56] L. Sjöberg, Emotions and risk perception, Risk Manag. 9 (2007) 223–237. [57] L. Sjöberg, Factors in risk perception, Risk Anal. 20 (1) (2000) 1–11. [58] S.H. Schwartz, Beyond Individualism/Collectivism: New Cultural Dimensions of Values, Sage Publication, Inc., 1994. [59] J.D. Sharp, M.K. Jaccard, D.W. Keith, Anticipating public attitudes toward underground CO2 storage, Int. J. Greenh. Gas Control 3 (5) (2009) 641–651. [60] P. Slovic, The Perception of Risk, Earthscan, London, 2000. [61] P. Slovic, M.L. Finucane, E. Peters, D.G. MacGregor, Risk as analysis and risk as feelings: some thoughts about affect reason, risk, and rationality, Risk Anal. 24 (2004) 311–322. [62] P. Slovic, M.L. Finucane, E. Peters, D.G. MacGregor, The affect heuristic, Eur. J. Oper. Res. 177 (2007) 1333–1352. [63] P.B. Smith, S. Dugan, F. Trompenaars, National culture and the values of organizational employees a dimensional analysis across 43 nations, J. Cross-Cult. Psychol. 27 (2) (1996) 231–264.

[64] Daniel Spreng, Gregg Marland, Alvin M. Weinberg, CO2 capture and storage: another Faustian Bargain? Energy Policy 35 (2) (2007) 850–854. [65] P. Sturgis, N. Allum, Science in society: re-evaluating the deficit model of public attitudes, Publ. Underst. Sci. 13 (1) (2004) 55–74. [66] Y. Takano, E. Osaka, An unsupported common view: comparing Japan and the U.S. on individualism/collectivism, Asian J. Soc. Psychol. 2 (1999) 311–341. [67] B.W. Terwel, Origins and consequences of public trust: towards an understanding of public acceptance of carbon dioxide capture and storage, in: PhD Dissertation, Kurt Lewin Institute, Leiden, 2009, dissertation series. [68] B.W. Terwel, F. Harinck, N. Ellemers, D.D. Daamen, Going beyond the properties of CO2 Capture and Storage (CCS) technology: how trust in stakeholders affects public acceptance of CCS, Int. J. Greenh. Gas Control 5 (2) (2011) 181–188. [69] B.W. Terwel, E. ter Mors, D.D. Daamen, It’s not only about safety: beliefs and attitudes of 811 local residents regarding a CCS Project in Barendrecht, Int. J. Greenh. Gas Control 9 (2012) 41–51. [70] M. Thompson, A. Wildavsky, A proposal to create a cultural theory of risk, in: H.C. Kunreuther, E.V. Ley (Eds.), The Risk Analysis Controversy: An Institutional Perspective, Springer-Verlag, Heidelberg, 1982, pp. 145–161. [71] K. Tokushige, K. Akimoto, T. Tomoda, Public perceptions on the acceptance of geological storage of carbon dioxide and information influencing the acceptance, Int. J. Greenh. Gas Control 1 (1) (2007) 101–112. [72] K. van Alphen, M.P. Hekkerta, W.C. Turkenburg, Accelerating the deployment of carbon capture and storage technologies by strengthening the innovation system, Int. J. Greenh. Gas Control 4 (2) (2010) 396–409. [73] K. van Alphen, P.M. Noothouta, M.P. Hekkerta, W.C. Turkenburg, Evaluating the development of carbon capture and storage technologies in the United States, Renew. Sustain. Energy Rev. 14 (3) (2010) 971–986. [74] K. van Alphen, J. van Ruijven, S. Kasa, M. Hekkert, W. Turkenburg, The performance of the norwegian carbon dioxide, capture and storage innovation system, Energy Policy 37 (1) (2009) 43–55. [75] K. van Alphen, Q. van Voorst tot Voorst, M.P. Hekkert, R.E. Smits, Societal acceptance of carbon capture and storage technologies, Energy Policy 35 (2007) 4368–4380. [76] S. Vercellia, S. Lombardi, CCS as part of a global cultural development for environmentally sustainable energy production, Energy Procedia 1 (2009) 4835–4841. [77] L. Wallquist, V.H. Visschers, M. Siegrist, Impact of knowledge and misconceptions on benefit and risk perception of CCS, Environ. Sci. Technol. 44 (17) (2010) 6557–6562. [78] L. Wallquist, S.L. Seigo, V.H. Visschers, M. Siegrist, Public acceptance of CCS system elements: a conjoint measurement, Int. J. Greenh. Gas Control 6 (2012) 77–83. [79] L. Wallquist, V.H. Visschers, S. Dohle, M. Siegrist, The role of convictions and trust for public protest potential in the case of carbon dioxide capture and storage (CCS), Hum. Ecol. Risk Assess. Int. J. 18 (4) (2012) 919–932. [80] M. Wang, P.S. Fischbeck, Evaluating lotteries, risks, and risk-mitigation programs, J. Risk Res. 11 (6) (2008) 775–795. [81] C. Welzel, How selfish are self-expression values? A civicness test, J. Cross-Cult. Psychol. 41 (2) (2010) 152–174. [82] E.U. Weber, C. Hsee, Cross-cultural differences in risk perception, but cross-cultural similarities in attitudes towards perceived risk, Manag. Sci. 9 (44) (1998) 1205–1217. [83] A. Wildavsky, K. Dake, Theories of risk perception: who fears what and why? Daedalus 119 (4) (1990) 41–60. [84] D. Williamson, Forward from a critique of Hofstede’s Model of National Culture, Hum. Relat. 55 (11) (2002) 1373–1395. [85] B. Wynne, Public engagement as a means of restoring public trust in science—hitting the notes, but missing the music? Commun. Genet. 9 (2006) 211–220. [86] A. Zhang, K. Moffat, J. Lacey, J. Wang, R. González, K. Uribe, L. Cui, Y. Dai, Understanding the social licence to operate of mining at the national scale: a comparative study of Australia, China and Chile, J. Clean. Prod. 108 (2015) 1063–1072.