Overcoming the headwinds: Can policy design shape public acceptance of wind power in Sweden?

This paper explores wind power attitudes in Sweden, considering the proximity of wind power installations, using a large-scale survey ( N = 5280). The study examines if attitudes were affected by policies that provide collective financial benefits through municipal tax revenues, personal benefits through direct compensation, or openings for democratic involvement. Only 15 % of the respondents expressed negative attitudes to wind power as a measure to speed up the transition to a fossil free society, while 26 % were negative to wind power built within 5 km from their homes. Attitudes were mainly predicted by ideological standpoints, environmental concern and political and governmental trust. The study found that for wind power constructed in the home municipality, respondents preferred collectively distributed financial benefits, while direct personal compensation offers the best prospects to influence ideologically motivated attitudes. None of the policy interventions tested in this study had any significant effect on respondents with strongly negative views.


Introduction
The development of cost-competitive low carbon energy technology, such a wind power, has made the prospects of achieving the target of keeping global warming well below two degrees Celsius, as committed to in the Paris Agreement, increasingly feasible.According to International Renewable Energy Agency, the global weighted-average cost of electricity from onshore wind projects fell by 69 % between 2010 and 2022 [1].Together with other clean energy technologies and more stringent climate policies, a transition away from fossil fuels by 2050 has become an increasingly realistic scenario [2,3].In effect, >117 countries agreed to triple renewable energy by 2030 at the COP28 in Dubai in 2023 [4].
The process of decarbonisation is however reliant of popular consent.In several democratic countries, renewable energy investments, particularly wind power installations, have caused political and social tensions [5,6].While public support for wind power expansion is high at national level in many countries, successful implementation of wind energy projects at local level is often low [7].Local protests against wind farms have become more common with the intensive expansion of wind energy [8,9], resulting in delays in the needed rollout of renewable energy.
Previous research has concluded that both the proximity of wind power sitings and the perception of distributional and procedural justice related to the construction process and operation of wind power are important determinants to explain support and resistance [10].While there is research exploring the social mechanisms behind acceptance, it is uncertain if individuals who express negative attitudes towards wind power project planned in their proximity can turn more supportive through compensatory measures, such as financial benefits, or by participatory planning processes.Furthermore, it is not fully known if such measures have different effects on the willingness to accept wind power within different social groups, and to what extent value-based factors, such as ideology, trust and environmental concern, affect acceptance attitudes and reduce the potential mediating effect of different compensatory measures or a participatory planning process.
This article aims to address these knowledge gaps by studying wind power acceptance in Sweden.Sweden is an interesting study case, being a country where wind energy has expanded rapidly and where the geographic conditions are favourable for further investments.Between 2001 and 2021, the installed effect increased by almost ten times [11].The expansion of wind power has, however, become an issue characterized by growing political tension in Sweden.Wind power is widely supported in society, yet attitudes have become increasingly polarized [12].
At the local level, wind power has become a particularly divisive issue.The Environmental Act and the municipal planning monopoly, entitle Swedish municipalities with the power to decline construction permits for wind power projects, colloquially referred to as the "municipal veto".The veto has turned municipalities into key decisionmakers on wind power licensing, and due to local level opposition, the number of municipal rejections has increased in recent years.In 2021, around 80 % of all construction applications were declined, leading to the situation when practically no new wind power projects are initiated [13].
The reasons for the trend with increasing rejections is not fully known, although it has been suggested that conflicts around land use and lack of local benefits contribute to the low acceptance rate [14][15][16][17].Sweden has not adopted any type of formal compensations for local communities living close to wind power farms or for host municipalities, which can partly explain why municipalities often turn down wind energy applications.Wind power developers are required to pay a minor property tax to the national government and land owners are compensated by lease agreements [16].People living nearby are, however, not entitled to any compensation, although some companies voluntarily pay community benefits, often used for rural homesteads or local projects.
In 2023, a governmental inquiry recommended the government to introduce incentives to affect local communities.One proposal was to enable municipalities to condition their approval of wind farm projects on the establishment of compensation agreements with local communities.
The inquiry suggested moreover that residents living within a radius of 10 times the total height of an onshore wind turbine should receive a certain percentage of the turbine's revenue during its lifetime [18].An issue of uncertainty, remains regarding what type of compensation would most effectively cultivate community acceptance.The inquiry report expressed that the type of compensation that would be most effective to address resistance, is to introduce a property tax that wind power operators would pay to the host municipality.Despite these proposals, no policies regulating compensation, or any other types of incentives, have been introduced in Sweden.
The Swedish case is not unique, and other countries are also struggling with similar public acceptance barriers.To successfully navigate a pathway towards low-carbon and sustainable future, there is a need for a deeper understanding of the social mechanisms behind support of different energy options.It is crucial to capture how various social and economic dimensions of the energy transition can be addressed and how different energy policies can be designed to cultivate support.
This article aims to contribute to these knowledge demands, by exploring attitudes and acceptance barriers to wind energy in Sweden.It analyses to what extent to which acceptance attitudes can be influenced by different policy designs, aimed at considering various fairness preferences.The paper begins with a literature review, followed by a theoretical discussion on the effectiveness of different compensatory models.It then outlines the research design and materials used.Finally, the results, discussion and conclusions are presented.

Explaining wind power acceptance
Wind energy can be contentious due to its visual and auditory impact.Several studies have shown that the physical intrusion of wind power installations can elicit negative reactions [8,9].The closer a wind farm is located to where people live, the less likely they are to accept it, a factor described as the proximity effect.This effect has shown to significantly influence on public opinions [19,20].Attitudes are also affected by factors such as the number and size of turbines [21,22] and place-attachment [23], although research findings on their impact on property prices are mixed [24,25].
While early studies on the public acceptance of wind power focused on the proximity effect [26,27], later studies have shown the location of the sittings cannot fully explain wind power acceptance and resistance [28,29].The characterization of wind power resistance as an ignorant and selfish reaction of nearby communities has thus been criticized [5,6].
Studies show that acceptance barriers can be associated with notions of being neglected or misrepresented [30] and perceptions of distributional injustice [31,32].Wind power does not necessarily generate any economic benefits for the local communities living near wind turbines, such as employment and financial returns [33].If the profits from energy generation are transferred to national level or to business corporations, individuals tend to see the construction of wind power as a form of exploitation of the local landscape [16][17][18].
People may therefore perceive that the economic outcomes derived from landscape usage are unfairly distributed between the affected communities living nearby, and the energy firms, often located far away.Several studies also demonstrate that individuals who economically benefit from wind power are more likely to accept its presence in nearby areas [34][35][36][37][38][39][40].There are also some research findings suggesting that acceptance of wind power is determined by personal norms, values and ideological orientations.Individuals who identify politically as leftleaning and prioritize environmental protection tend to be more positive to wind power [41,42].
The association between ideology and public attitudes towards different energy technologies can also be influenced by political narratives [43] and by media messaging and political debates [44][45][46].Distinguishing personal values from the impact of political narrative can be challenging.Some studies have also indicated that attitudes can be manipulated by economic agents or interest groups advocating against for or against the usage of renewable energy [47].

How to overcome public acceptance barriers?
To cultivate acceptance for wind power, governments and energy firms have implemented different forms of compensation or community benefits.Wind power operators may, on voluntary or legal basis, distribute financial compensation, facilitate community co-ownership, support the local economy or civil society, or offer discounted electricity prices [48].In some countries, operators are mandated to pay local property or corporate taxes to host municipalities [49].
There are studies suggesting that financial compensations may positively influence public acceptance of wind power projects [21,[50][51][52], although the effect varies depending on contextual factors such as the level of compensation, and the distance to, size and number of wind power turbines [53].Conversely, other studies indicate that compensations might strengthen opposition if they are perceived as bribes [54][55][56].The distribution can be seen as unfair and the compensation deemed insufficient [22,57,58].
Moreover, research suggests that the decision-making process and the ability of affected individuals to participate in the planning process of wind power projects are relevant factors for acceptance [59,60,61].Individuals who distrust public institutions or perceive the decisionmaking processes as unfair, non-transparent or illegitimate, are also more likely to oppose wind farms [62].These findings underscore the relevance of perceptions of procedural injustice in understanding why certain individuals express negative attitudes to wind power installations.

Theorising acceptance barriers to wind power
The objective of this paper is to deepen the understanding of public D. Lindvall et al. attitudes towards wind power by exploring if different forms of financial benefits and openings for democratic participation can enhance support.
First of all, attitudes to wind power vary depending on factors such as ideological orientations, environmental concern, and political and governmental trust [22,43].It remains uncertain, however, how significantly personal values influence attitudes to wind power development, and whether this influence varies depending on the proximity of the wind power project and the hierarchy between the value-based aspects (e.g., whether certain values are more influential than others in explaining wind power support).
To respond to these uncertainties, we explored the following research question: R1.To what extent is the support of wind power predicated by valuebased factors (ideological orientations, environmental concern, political and governmental trust) or sociodemographic factors (e.g.residence, income, age, gender)?
There are a growing number of studies concluding that the perception of distributional and procedural fairness are key determinates for wind power support [22,53].It can accordingly be assumed that individuals who receive financial benefits or who are given an opportunity to participate in and influence the decision-making process, are more likely to support wind power projects.
Addressing fairness perceptions with financial compensation can be challenging, given that fairness is a multifaceted concept.Individuals might, for instance, emphasise different distributional norms, such as equity or equality [63].These principles can be expressed in terms of personal or collective compensation.While studies on climate policy acceptance have shown that people can be guided by these two different fairness preferences [64], it is uncertain which of these two types of compensation most effectively can address perceptions of unfairness in the context of wind power projects, and whether different social segments express similar fairness preferences.
It can, on the one hand, be assumed that individuals living near a wind project who emphasise equity as a distributional norm would prefer compensation distributed on a personal basis, and thus be affected by direct financial compensation.Individuals who emphasise equality, should instead opt for compensation distributed collectively, for instance by requiring the wind power investor to pay taxes to the host municipality.This would enable the municipality to increase social spending to the benefits of the wider society.
Previous studies have suggested that the proximity of wind power influences preferences of compensation, and that the closer individuals live to a wind turbine, the less they prefer compensation to be collectively distributed, while groups who are not living in the direct vicinity of the wind power sittings, favor collective solutions [65].
Finally, avoiding perceptions of procedural injustice by inviting individuals affected by the project to participate in decision-making processes is likely to increase support for wind power investments.
To explore these associations, the following research question is asked: R2.To what extent can policy measures, addressing distributional outcomes and procedural aspects, influence the support for wind power projects in the proximity of the individual.
Finally, it remains unclear how different sociodemographic factors such as income and place of residence (rural/urban), and value-based factors such as ideology, environmental concern, and trust, predict fairness preferences in the context of wind power investments.These factors may also influence the effectiveness of different policies addressing distributional outcomes and procedural outcomes.
As demonstrated in the literature review, personal values significantly influence wind power attitudes [42], yet it is uncertain to what extent these factors are influencing the effect of financial benefits and procedural process on wind power support.
These uncertainties motivate us to ask: R3.To what extent are sociodemographic factors (e.g.residence, income, age, gender) and value-based aspects (e.g., ideology, environmental concern, and political and governmental trust), influencing the effect of policies addressing personal or collective distributional outcomes and procedural aspects, on support of wind power projects in the proximity individuals.

Methodological approach
The results of this study are drawn from a survey [66], where respondents were queried about their general attitudes towards wind power development as a mean to accelerate the transition to a fossil free society.Responses were captured using a scale ranging from one to seven, with four being a neutral category.Immediately following the initial question, respondents were asked whether they would support the construction of wind power in their own municipality and whether they would support construction of wind power at a distance of less than five kilometres from their home.
The choice of maximum distance is partly based on the proposal of the government inquiry, which suggested compensation rights for individuals living in close proximity to wind power turbines [18], calculated on the height of the wind turbine.To ensure an objective measurement, we chose for 5 km, as this aligns with the maximum distance used in the national strategy for sustainable wind power development of the Swedish Energy Agency, for assessing the degree of conflict and the possibility of coexistence with other land use interest [67].Additionally, the regional authorities advise developers against constructing wind power within 5 km of peoples' homes [68].
After responding to each of the last two questions, respectively, participants were randomly assigned to one of three experimental conditions.In each condition, participants were asked if they would be more supportive to wind power construction under different scenarios: if they were receiving financial benefits, distributed collectively as a tax to the host municipality, personally as a compensation or discounted electricity, or provided with the opportunity for democratic influence over

Table 1
Formulations of baseline and alternative survey questions.

Initial questions
Questions with attributes To speed up the transition to a fossil free society, I support the construction of more wind power in Sweden.I would support the construction of wind power in my municipality.I would be more supportive to the construction of wind power in my municipality… …if the operator would pay tax to the municipality, enabling increased funding of for example schooling, health and public care services.…if I would first be given real possibility to influence the decision to construct wind power in my municipality.…if I as a resident in the municipality would receive compensation or discounted electricity?I would support the construction of wind power <5 km from my house.I would support the construction of wind power <5 km from my house… …if the operator would pay tax to the municipality, enabling increased funding of for example schooling, health and public care services.…if I would first be given real possibility to influence the decision to construct wind power near where I live.…if I as a person living nearby would receive compensation or discounted electricity?
D. Lindvall et al.
the decision (see Table 1).
The first two alternatives represent collective and personal compensations, addressing two aspects of distributional justice, while the third represents procedural justice.These three alternatives were also aligned with the proposal of the government inquiry, which advocated for a system involving both direct compensation to individuals living near wind turbines and a tax for the host municipality.The inquiry also emphasized the importance of participation and democratic influence for affected communities [18].
This methodological design allowed us to conduct a comparative analysis between the baseline attitudes and the different policy designs, as well as to test to if the alternative policy designs could turn negative baseline attitudes to wind power into more supportive attitudes.To this end, we conducted two sets of analysis.
First, we compared the different models of compensation, and participation, on each level of proximity, assessing which one of the three was rated most highly.Second, we examined the negative baseline responses, with grades 1 to 3, and assessed the extent to which the conditional questions could bring these responses upwards.This analysis provides insights into the effectiveness of the three policy interventions.
Our objective was to assess the effect of the interventions over the initial question.We decided not to employ a within-subjects design and present all conditional questions to all respondents because we wanted to avoid progressive effects from asking the all three questions to each participant.Instead, each respondent replied to the baseline question, and to one conditional question for each level of proximity in a betweensubjects design.
The large representative sample size and the randomization of participants across the three conditions asserted the appropriateness of using a between-subjects design (i.e., randomly assigning each participant to one of the three experimental conditions).In the final section of the questionnaire, we collected data on sociodemographic factors, such as income, rural/urban residence, gender, and age, along with responses on value-based factors, including ideological orientation, environmental concern and trust in governments and in politicians.
To measure environmental concern, we utilized a set of questions adapted from a scale developed by Schultz [69].Respondents were asked to assess their concerns about environmental impacts across various dimensions: themselves, all people, their lifestyle, people close to them, animals and plants, future generations, and children.Based on their responses, we constructed an index to quantify the level of environmental concern.
We chose to examine ideological leanings on both a left/right scale and a scale capturing GAL (Green/Alternative/Libertarian) and TAN (Traditional/Authoritarian/Nationalistic) values.The GAL-TAN scale encompasses attitudes towards multiculturalism, gender equality, and international solidarity on one end, and conservatism and national community on the other [70].GAL/TAN values are typically assessed using a set of value judgments [71,72].We constructed a GAL/TAN index by calculating the mean responses to the questions in Table 2.For a comprehensive overview of the survey design, please refer to the supplementary material.
The results were analyzed using both descriptive and inferential statistics.Multiple-regression analyses were employed to explore how policy acceptability is underpinned by individual differences in sociodemographic variables and value-based factors.The multiple-regression analyses did not fully meet a strict interpretation of the underlying assumptions of normally distributed variables and absence of multicollinearity.Because of this, the goodness of fit estimate should be taken with caution.The main purpose of the multiple-regression analyses was, however, to compare the relative strength of the predictors.

Data sample
The survey was distributed via paper to Swedish residents aged 18 to 84 and was conducted between April 25 and August 23, 2023, by the company SKOP.In addition to the questions analyzed in this research, the survey included inquiries about other climate policies that were not analyzed in this study.Participants were randomly chosen from the population register, making this survey unique as it did not utilize web panels.Respondents were selected through a non-proportional stratified sampling method across three age categories and divisions between different types of municipalities to ensure proportional representation.
The respondents had the option to complete the survey either on paper or in a web format.Data collection was conducted in three waves to obtain a sufficient number of responses.The response rate for the first wave was 21 %, while for the final two waves, it was 16 %.The total sample comprised 33,432 respondents.Post-stratification was then applied to ensure the sample was representative of the total population.After weighting the dataset based on several core demographics (municipality groups, gender, and age) to align with the Swedish population, a total of 5280 respondents were included in the dataset.
Following the post-stratification processes, the dataset was fairly well balanced and representative of the Swedish population, comprising of 48.8 % women and 50.6 % men and 48.7 % living in an urban area and 51.3 % in a rural area.The reported average income was 45.85 thousand SEK (SD = 68.26)and the average age was 54.64 years (SD = 17.32).Although the average income and age was slightly higher than the national averages, these marginal differences were not deemed to significantly affect the representativeness of the sample.The observations were not adjusted in any way.The data included in our statistical analyses are thereby the data from the weighted sample as it were when collected.The reason why the sample varies in some analyses, is because data are missing in some few instances, and because some respondents did not respond to all questions in the questionnaire.Further details about the data collection and stratification methodology are provided in the supplementary material accompanying this article.

General support for wind power development for speeding up the transition to a fossil free society
The respondents were generally positive towards the development of new wind power, as a way to speed up the transition to a fossil free society.However, they were less positive to wind power development in their home municipality or at a location near their own home (Table 3).
As can be seen in Table 4, a large portion of respondents were highly supportive of more wind power (i.e., responding 7 on a scale ranging from 1 to 7), while a small group strongly opposed it (i.e., answering 1 Libertarian/ alternative It is important to have an environmentally sustainable society.

Green
In a secure society, immigration is low, and the majority of the inhabitants share the same culture.
Traditional/ nationalist In a secure society, law and order is maintained by a strong police force and a judiciary that can sentence cognizable punishments for criminals.

Authoritarian
Where do you place yourself politically on a right-left political spectrum?
Right/left Swedish public authorities are mostly acting proficiently, impartially and in accordance with the law.

Governmental trust
Swedish politicians are mostly acting in the interest of the public.

Political trust
Note: For GAL we calculated the average value of 1-7 responses to the questions on Green and Libertarian/Alternative, and similarly on the questions regarding Traditional/Nationalist and Authoritarian.
D. Lindvall et al.
on a scale ranging from 1 to 7).Of note, few participants were slightly negative.Among the negative participants (i.e., those who responded 1-3 to the questions), most were in strong disagreement.Furthermore, participants were significantly more likely to express very negative attitudes (i.e., responding 1 on the scale ranging from 1 to 7) towards wind power development 5 km from home versus wind power development in home municipality, as shown in a proportion comparison test, z = 8.58, 95 % CI [0.04-0.07].Similarly, participants were significant more likely to be very positive (i.e., responding 7 on the scale ranging from 1 to 7) towards wind power development in home municipality versus wind power development within 5 km from home, z = 11.93,95 % CI [0.09-0.13].
Table 5 reports the means for the predictor variables.A regression analysis was conducted with general support of wind power as a way to speed up the transition to a fossil free society being uses as the dependent variable.Environmental concern, political ideology, trust in governments, trust in politicians, GAL and TAN were included as independent variables to test how wind power support depends on these predictors individually.Living conditions (48.7 % urban vs. 51.3% rural), monthly income (M = 45.85 thousand SEK, SD = 68.26),gender (48.3 % female, 50.2 % male, and 0.6 % other), age (M = 54.65 years, SD = 17.32, range 18-85) and length of education (M = 2.98, SD = 1.02, range 1-4) were also included in the analysis as control variables.
As can be seen in Table 6, all variables contributed with a significant explanation of unique variance in wind power support, except the two control variables gender and income.In response to research question 1, the results revealed that the two strongest predictors on wind power support were GAL and environmental concern, followed by political orientation, trust in governments, and trust in politicians.The living condition control variable also contributed significantly, with a higher willingness to support more wind power in urban than in rural areas.The regression model explained approximately 36 % of the total variance, R 2 = 0.36, R = 0.60, p < .001.
We also explored the characteristics of the individuals (17 %) who expressed strong negative attitudes to wind power within 5 km from home (Table 7).Compared to the sample average, these individuals are characterized by a relatively low environmental concern, being slightly more right-wing in their political orientation, have a lower trust in governments, a slightly lower trust in politicians, being less GAL and more TAN.
We also explored the attitudes to new wind power among residents in areas with high versus low concentrations of wind power, by studying answers of respondents living in the 19 municipalities with >200 MW installed capacity, representing around 60 % of the total installed wind power in Sweden, and comparing them with respondents living in other municipalities.We were, however, unable to observe any differences between respondents living in the top-producing municipalities, and those living in other municipalities.The Bayes factor for the difference between participants living in top-producing municipalities and participants living in other municipalities for attitudes to more wind power generally was 21.44 (a Bayes factor suggesting extreme evidence for the null-hypothesis, i.e., no difference between the two groups).For attitudes to more wind power in the home municipality the Bayes factor was 6.87 (a Bayes factor suggesting strong evidence for the null-hypothesis,

Table 3
Mean response to questions concerning a general support for more wind power plants being built in Sweden, in the home municipality, and within 5 km of the respondents' own home.Note: estimates for each question, respectively, were made on a scale ranging from 1 to 7. The standard deviations in parentheses.

Table 4
Frequency of responses for each response category for questions concerning general support of more wind power plants being built in Sweden, in the home municipality, and within 5 km of the respondents' own home.Note: The responses were made on a scale ranging from 1 (not supporting at all) to 7 (very high support).

Table 5
The means and standard deviations (SD) of the predictor variables in the study.Note: R2 = 0.36, R = 0.60.

Table 7
The means and standard deviations (SD) of personal characteristics of respondents strongly negative to wind power within 5 km of their own home.i.e., no difference between the two groups), and for attitudes to more wind power closer than 5 km from respondents' home it was 2.37 (a Bayes factor suggesting that the null-hypothesis is about 2 times more likely than the hypothesis, i.e., no difference between the two groups).

Effects of policy design on support of wind power development
Table 8 presents the difference between respondents' tendency to be supportive of new wind power development in their own municipality give that the development is associated with collective, procedural, or personal justice consequences.An estimate at 4 represents "no change", whereas an estimate above 4 represents "more supportive" and an estimate below 4 would represent "not more supportive" wind power.An estimate of 1 represents "definitely not more supportive" and an estimate of 7 represents "definitely more supportive".
The findings revealed that respondents were overall more supportive of wind power development in their own municipality by all policy designs, the largest effect found in the policy with a collective consequence.Participants who were positive to new wind power in Sweden generally (those who answered 5-7 on the baseline question) would be even more supportive if the development had desirable consequences, in particular if it had a collective consequence.In turn, participants who were negative to new wind power in Sweden as a general transformative measure (those who answered 1-3 on the baseline question) did not turn more supportive after being presented with policies with justice consequences.When only respondents initially strongly negative (those who answered 1 on the baseline question) were included in the analysis, the results revealed that this group was clearly not more supportive even if the development was associated with certain forms of justice.
A similar pattern was found for wind power development at a location <5 km distance from the respondents' own home (Table 9).Here, participants were overall less supportive and similarly unconvinced by the policy designs.Again, participants who were supportive at baseline were also the ones who reported an even more positive attitude to wind power if the development was associated with certain forms of justice.Those who were initially negative (1-3) or strongly negative (1) to wind power, in turn, reported being no more positive regardless of the type of policy design.In response to research question 3, we can conclude that the policy designs used in the current study had little effect on support of new wind power in the respondents' proximity.A collective justice policy appears to be the most effective in making people more positive to wind power development, but those who are negative to new wind power are generally not turned by that justice type.
Next, we conducted a more fine-tuned analysis of participants' susceptibility to become more supportive to wind power development and analyzed how the group with negative (1-3) and participants strongly negative views (1) (at baseline) were distributed across responses to the policy designs.Even if not shown in the mean analyses (Table 7), some negative participants were more supportive of new wind power in their municipality, especially if the development would have collective distributional justice (Table 10).Nearly 25 % of the negative participants would be more supportive following this type of policy treatment.
Attitudes related to new wind power development at a location <5 km from the respondents' own home were even more robust (Table 11).About 15 % of the negative participants were more supportive of new wind power near their home following a collective policy treatment, and even fewer following a procedural or personal justice policy treatment.

Factors contributing to estimates of increased willingness to support new wind power development by fair policy treatment
In a first set of analyses, we tested how individual variations in environmental concern, political ideology, trust in governments, trust in politicians, GAL and TAN, and the control variables living condition, monthly income, and gender, contributes to the explanation support for development of new wind power plants in the respondents' municipality.Separate regression models were conducted with increased support for new wind power in own municipality as dependent variable, after collective procedural and personal distributional policy treatment (Table 12).
Environmental concern consistently emerges as the strongest support for new wind development in respondents' home municipality across all three types of policy treatment.Political ideology also contributes similarly across all treatments.The more left-wing oriented participant, the stronger the effect of the policy design on estimates of higher support for new wind power.Differences can also be noted.Higher GAL scores are associated with a higher increase in support for new wind power, but only in the collective and the procedural policy design conditions, not in the personal policy design condition.Similarly, higher trust in politicians had a positive association with higher increase in support of new Note: Increased policy acceptance estimates were made on a scale from 1 to 7 where 4 represented a "neutral" standpoint.Thus, means above 4 represent a positive evaluation of the policy and means below 4 represent a negative evaluation of the policy.Note: Increased policy acceptance estimates were made on a scale from 1 to 7 where 4 represented a "neutral" standpoint.Thus, means above 4 represent a positive evaluation of the policy and means below 4 represent a negative evaluation of the policy.
wind power, but only in the procedural and the personal policy design conditions, not in the collective policy design condition.Finally, living conditions (urban vs. rural) only had an effect in the personal policy design condition, where individuals living in rural areas tend to be more positively influenced by the policy design.Lastly, separate regression models were conducted with increased support for new wind power <5 km from home as dependent variable, after collective, procedural and personal distributional policy treatment (Table 13).Again, environmental concern comes out as the strongest predictor of higher support for new wind development within 5 km of respondents' home across all three types of policy treatment.Political ideology as well as GAL values also contributes similarly across all treatments.The effect of policy designs emphasizing collective and personal fairness framing were strongly determining support for new wind power for participants who are left-wing oriented and had higher GAL score.
Differences can also be noted.Interestingly, higher TAN scores and gender (men) were associated with a larger increase in the positive attitude to the policy when the development had personal financial gains, but not otherwise.Trust in politicians contributed to collective framing, but only under this specific condition.In sum, we can conclude, in response to research question 4, that people's tendency to become  Note: Baseline response refers to participants' attitude to more wind power in Sweden generally.

Table 12
Results from a multiple regression model with support for new wind power in own municipality following collective, procedural and personal policy treatment as dependent variables.
Predictor more positive towards new development of wind power is primarily determined by a concern for the environment and ideological standpoints (political left-right orientation and green-alternative-libertarian, GAL).

Discussion
Negative attitudes to wind power, both generally and at local level, pose a major challenge to the climate and energy transition.Therefore, reaching deeper understanding of what it takes to overcome acceptance barriers are of great relevance.Contrary to the negative wind power rhetoric expressed in the political debate [7], this study demonstrates that the support for wind power in Sweden is strong.More than threefourths of the respondents had a generally positive attitude to wind power development.
It is possible that the attitudes were influenced by the phrasing of the question, asking respondents about their support for wind power as a way to accelerate the transition to a fossil free society, and that they would be less supportive of wind power in general.The survey reveals, however, that support is strong for wind power in the respondents' home municipality and that the respondents were generally positive towards wind power built near their home.Nevertheless, the findings suggest that there is small minority who are very negative, and opinions in general are either strongly in favor or strongly against wind power.
Regarding the first research question, the results suggest, in line with previous findings [42,43], that value-based aspects, particularly environmental concern, ideology and political and governmental trust, can explain positive or negative attitudes towards wind power.Individuals who are environmentally concerned and who express left-wing, green, libertarian and alternative (GAL) views are among the most supportive of wind power.In contrast, those with low environmental concern, low trust and right leaning and traditional, authoritative and nationalistic (TAN) orientations, tend to hold negative opinions..There is also an urban and rural divide, with urban residents being more supportive, yet rural residence is not a key determinate for negative attitudes.The identified R2 values in our regression analysis were however relatively low, suggesting that the explanatory power of our models is somewhat limited.
Another conclusion that can be drawn from the study, in relation to the second research question, is that policies that provide financial benefits for affected communities or individuals, or offer the ability to exert democratic influence, have marginal effects on negative respondents.Very few respondents with strongly negative baseline views became more supportive after being exposed to policies with any of the three justice framings tested in this study.In this regard, the findings do not support previous research, suggesting that resistance should first of all be understood as a matter of perceived distributional or procedural injustice [22,53,62].This discrepancy can perhaps be explained by the research design, which captures attitudes expressed by individuals who may not have any personal experience of wind power being built near their home.
It is also possible that the politization and polarization of energy policies in Sweden have affected opinions, and it can thus be expected that wind power attitudes are shaped by politically motivated reasoning [73].Respondents may simply have a psychological tendency to adjust their attitudes to be more lenient towards those expressed by members of the community to which they consider themselves to belong, while being more critical and less forgiving of the opinions of those outside that community.
Although the empirical findings do not allow firm conclusions about the reasons behind these attitudes, it is possible that negative respondents view wind power as a political concept, and are unlikely to be swayed by financial benefits or openings for democratic participation.Wind power attitudes may also be affected by political cues [43] and it is likely that individuals who express strong TAN values are more receptive to political narratives questioning wind power.This stance is currently pursued by many conservative and right-wing populist politicians in Sweden [16].
This finding does not imply that perceptions of unfairness are irrelevant for wind power acceptance in Sweden.First, all respondents became more supportive after being treated with the alternative policy options.The preference for collective justice framing indicates that many respondents consider compensation to municipalities hosting wind power turbines as a matter of fairness.The study indicates moreover that some individuals with negative attitudes towards wind power became more supportive after being exposed to the policy alternatives explored in this study, although the effect was marginal.This suggests that wind power installments will upset the attitudes of some individuals, regardless of the financial benefits offered.
Regarding research question three, the study found that fairness preferences expressed by respondents differed depending on valuebased standpoints.Unsurprisingly, left-leaning individuals are more prone to support collectively distributed benefits, which was also the case for individuals expressing strong GAL-attitudes.Those living in rural areas, on the other hand, preferred personal compensation if wind power is constructed in their home municipality.This preference perhaps indicates that they are more likely to be personally affected by wind power installation and thus wanting to be personally compensated, as has also previously been suggested [66].
When we explored wind power projects near the homes of respondents, we found that those who expressed negative attitudes were generally more supportive of personal compensation.This was particularly evident among individuals who expressed TAN-associated values.Previous studies have also demonstrated that individuals who express self-enhancing values tend to be less supportive of progressive carbon taxation [74,75], making it reasonable to believe that such attitudes also correspond with preference for personal compensation in the context of wind power development.Moreover, the results demonstrate that individuals who express left-wing attitudes and high political trust, prefer compensation to be collectively distributed through the municipality if the project is close to their home.Fairness preferences seem to differ accordingly, depending on ideology, worldviews and political trust.

Conclusions and policy-implications
The findings of the study suggest that overcoming wind power resistance in Sweden is challenging, given that negative attitudes are associated with ideological standpoints, worldviews and trust.Wind power development is likely to remain contested, regardless of policy interventions providing financial benefits or ensuring democratic participation.However, this conclusion does not suggest that such policies are entirely ineffective.By distributing revenues to host municipalities, providing individual compensation or ensuring democratic participation in the decision-making process, individuals could become more supportive of wind power projects.Additionally, such policies could influence local politicians who make final decisions of project applications.
The study also indicates that distributing personal financial benefits has the best prospect for influencing attitudes of individuals living near wind power installations.However, the effect is marginal, and such policies cannot be considered as a silver bullet against negative attitudes to wind power.Moreover, environmental concern consistently emerges as the strongest predictor of support for new wind development across all three types of policy treatments examined here.Policymakers are accordingly advised to consider a combination of policy interventions to influence different groups.
There are limitations to the conclusions that can be drawn from a survey-based study on hypothetical wind power projects.Most respondents have no personal experience of wind power being constructed close to their homes or the policy alternatives presented in this study.It is reasonable to believe that individuals who express supportive attitudes towards hypothetical projects, might become more skeptical if a project is actually commenced nearby their homes.Conversely, those whose negative attitudes are shaped by values and ideological standpoints could change their minds after having received financial benefits or having been able to participate in and influence the decision-making process.Further research should thus be conducted to explore attitudes of communities with actual experience of wind power siting, perhaps by using geodata to identify individuals living close to wind power turbines.

Ethical review
The survey, the methods used for the data collection and storage has been approved by the Swedish Ethical Review Authority (Dnr 2023-01099-01).

Declaration of competing interest
We have no conflicts of interest to disclose.

Table 2
Formulations of survey questions on ideology and worldviews.

Table 6
Multiple regression model with support for new wind power in Sweden as dependent variable.

Table 8
Increased support for wind power development in home municipality if the implementation of the policy would have collective, procedural or personal justice consequences.

Table 9
Increased support for wind power development within 5 km of respondents' home if the implementation of the policy would have collective, procedural or personal justice consequences.

Table 10
Impact of justice framings on respondents expressing negative attitudes to more wind power in municipality.

Table 11
Impact of justice framings on respondents expressing negative attitudes to more wind power within 5 km of home.

Table 13
Results from a multiple regression model with support for new wind power within 5 km of own home following collective, procedural and personal policy treatment as dependent variables.Flags statistically significant values.Degrees of freedom and goodness of fit estimates: For collective justice (df = 1422, R 2 = 0.15), for procedural justice (df = 1502, R2 = 0.09) for personal justice (df = 1501, R2 = 0.07).
a D.Lindvall et al.