How big is my backyard - A survey and field experimental study on the social acceptance of Agri-Photovoltaics
How to cite
Fesenfeld, Lukas; Schmidt, Tobias; Rohrer, Jürg; Anderegg, Dionis; Montfort, Simon; Wehrli, Stefan; Aschemann-Witzel, Jessica; Torma, Gabriele. 2024. "How big is my backyard - A survey and field experimental study on the social acceptance of Agri-Photovoltaics." AEA RCT Registry. AEARCTR-0013540. https://doi.org/10.1257/rct.13540-1.1
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
To reduce the carbon footprint of our society, deployment of renewable energies needs to be acclerated. At the same time, carbon intensive energy production plants as well as nuclear power plants are being phased out. As a consequence, the installation of open-space and utility scale solar energy (USSE) plants has been promoted in the last years. Given that such solar PV installations often have a larger land-use footprint than other power plants (Trainor et al., 2016), the deployment of solar PV increasingly competes for limited land resources. This is particularly pronounced in regions like Switzerland, where land resources are limited and the landscape awareness is high (Huber et al., 2017). The current political discussion in Switzerland about open-space and utility scale solar energy shows that expanding open-space solar installations is often seen as competition to other land-use purposes like food production. In this context, Switzerland's degree of food production self-sufficiency is used by some actors as an argument against open-space solar installations. The emergence of Agri-Photovoltaics (Agri-PV), the co-utilization of land for both agricultural purposes and solar energy generation, offers a potential solution to that conflict by maximizing land productivity and resource efficiency (Trommsdorff et al., 2021). However, the successful integration of Agri-PV systems hinges not only on technological feasibility but also on social acceptance (Pascaris at al., 2021).
Local opposition to energy projects is often referred to as a not-in-my-backyard (NIMBY) problem, a phenomenon where individuals or communities express opposition to the siting of undesirable facilities, such as energy infrastructure, in their vicinity. This is the case even though individuals might generally support the deployment of wind and solar in their country (Boyle et al., 2019). Research has shown that the NIMBY phenomenon is overly simplistic and that next to proximity, there are various other factors that influence local support for energy projects. These include place attachment, visual impact and social norms (Devine-Wright, 2004; Liebe & Dobers, 2019).
Nevertheless, a growing body of research consistently shows that distance to (planned) renewable energy projects can play an important role in mobilizing opposition against solar- and wind projects. For instance, in a nationally representative survey in Germany, Bertsch at al. (2016) found out that while the majority of Germans approve the expansion of RET and the power grid, local opposition is mostly fueled by concerns regarding the landscape modifications this entails. Additionally, the distance of energy infrastructure from the place of residence seems to have a strong influence on the RET project acceptance of local residents. Bertsch et al. found out that solar PV-modules installed 1000m away from places of residence are still supported by 85% of residents. For installations closer than that, support drops significantly. At any distance, support for wind energy was lower than support for solar PV. This is consistent with results from Schumacher et al. (2019) where support for wind and solar projects increased with increasing distance to the places of residents. While Bertsch et al. (2016) did not specify the size of solar PV projects, Schumacher et al. (2019) differentiated between small-scale and large-scale solar projects. The distance-dependent support for small-scale solar projects was comparable to the results of Bertsch et al. (2016) while the support for large-scale solar projects was considerably lower. Moreover, in Ontario, a study shows that citizens punished the incumbent liberal party during provincial elections for nearby wind energy projects. While wind energy proposals within 3km of their precinct were associated with a 5% decline in vote shares, operational wind turbines were associated with a 10% decline in vote shares (Stokes, 2015).
Even though research shows that Agri-PV might increase public support for local solar development (Pascaris et al., 2022), the influence of potential NIMBY effects and the proximity of (planned) Agri-PV projects on citizens' support for Agri-PV and renewable energy policies has not yet been examined. In addition, to the best of our knowledge, potential NIMBY effects on revealed voting behaviors of citizens has not yet been studied in a randomized-controlled field experiment. This study aims to close these research gaps.
In addition, as support for renewable energy technologies is not only influenced by proximity but also by a multitude of other factors (Devine-Wright, 2004; Liebe & Dobers, 2019), this study further investigates support on potential Agri-PV projects depending on different project characteristics, like the Agri-PV type, size, ownership structure, distance, impact on communities’ own production with food, impact on communities’ own production with energy, and project’s impact on farmer income. Based on a review of the existing literature, we formulated the following research questions.
Research Questions:
1. How does information about the realistic potential (small or large) for developing Agri-PV projects in the proximity to citizens’ place of residence (i.e., 0-500 meter, 500-1500 meter, and 1500-4500 meter around individuals’ homes) affect citizens’ attitudes towards Agri-PV, their support for Agri-PV projects, their policy support for renewable energies and their intended and revealed voting behavior?
2. How do different characteristics of potential Agri-PV projects (i.e., Agri-PV type, size, ownership, distance to place of residence, impact on communities’ own production with food, impact on communities’ own production with energy, and impact on farmer income) affect citizens’ Agri-PV project support?
To test these questions, we run a field- and survey-experiment with a representative sample of Swiss citizens and permanent residents (+18) years in the French and German speaking part of Switzerland. The sample is provided by the Swiss Federal Statistical Office. We conduct two survey waves. The first wave is conducted in early May 2024 just before the direct democratic vote on renewable energies on the 9th of June 2024 (the referendum on the so-called Electricity law that aims to accelerate the installation of renewable energy, especially open-space solar PV and wind plants in Switzerland). As part of the first wave, we randomly treat a representative subsample of the respondents with information about the realistic potential of developing Agri-PV plants in the close proximity of their place of residence. We also include various items on citizens’ attitudes towards Agri-PV, their policy support, and intended voting behavior during the upcoming referendum on the Electricity law into the first survey wave. Moreover, participants conduct a conjoint experiment, in which we randomly vary relevant Agri-PV project design features. In the second survey wave, after the direct democratic vote on the 9th of June, we then again ask respondents that participated in the first survey wave on their attitudes towards Agri-PV, their policy support, and their actual voting behavior during the referendum on the Electricity law. This setting allows us to draw causal inferences about potential NIMBY effects and the research questions outlined above.
Intervention
We conduct two survey waves. The first wave is conducted in early May 2024 just before the direct democratic vote on renewable energies on the 9th of June 2024 (the referendum on the so-called Electricity law that aims to accelerate the installation of renewable energy, especially open-space solar PV and wind plants in Switzerland). As part of the first wave, we randomly treat a representative subsample of the respondents with information about the realistic potential of developing Agri-PV plants in the close proximity of their place of residence. We also include various items on citizens’ attitudes towards Agri-PV, their policy support, and intended voting behavior during the upcoming referendum on the Electricity law into the first survey wave. Moreover, participants conduct a conjoint experiment, in which we randomly vary relevant Agri-PV project design features (see Experimental design section for details). In the second survey wave, after the direct democratic vote on the 9th of June, we then again ask respondents that participated in the first survey wave on their attitudes towards Agri-PV, their policy support, and their actual voting behavior during the referendum on the Electricity law. This setting allows us to draw causal inferences about potential NIMBY effects and the research questions outlined above.
Based on the existing literature, we developed the following hypotheses:
H1: Information about the realistic potential (small or large) for developing Agri-PV projects in the proximity to citizens’ place of residence (i.e., 0-500 meter, 500-1500 meter, and 1500-4500 meter around individuals’ homes) affects citizens' attitudes towards Agri-PV.
H1a: Information about a large potential for developing Agri-PV projects in closer distance to citizens’ place of residence (so 0-500 meter compared to 500-1500 meter or 1500-4500 meter) reduces citizens’ attitudes towards Agri-PV compared to citizens’ provided with no such information or with information about a small Agri-PV developing potential in citizens’ close proximity.
H2: Information about the realistic potential (small or large) for developing Agri-PV projects in the proximity to citizens’ place of residence (i.e., 0-500 meter, 500-1500 meter, and 1500-4500 meter around individuals’ homes) affects citizens’ Agri-PV project support.
H2a: Information about a large potential for developing Agri-PV projects in closer distance to citizens’ place of residence (so 0-500 meter compared to 500-1500 meter or 1500-4500 meter) reduces citizens’ Agri-PV project support compared to citizens’ provided with no such information or with information about a small Agri-PV developing potential in citizens’ close proximity.
H3: Information about the realistic potential (small or large) for developing Agri-PV projects in the proximity to citizens’ place of residence (i.e., 0-500 meter, 500-1500 meter, and 1500-4500 meter around individuals’ homes) affects citizens' policy support for renewable energies.
H3a: Information about a large potential for developing Agri-PV projects in closer distance to citizens’ place of residence (so 0-500 meter compared to 500-1500 meter or 1500-4500 meter) reduces citizens' policy support for renewable energies compared to citizens’ provided with no such information or with information about a small Agri-PV developing potential in citizens’ close proximity.
H4: Information about the realistic potential (small or large) for developing Agri-PV projects in the proximity to citizens’ place of residence (i.e., 0-500 meter, 500-1500 meter, and 1500-4500 meter around individuals’ homes) affects citizens' voting behavior in a direct democratic vote on renewable energies (the so-called Electricity law).
H4a: Information about a large potential for developing Agri-PV projects in closer distance to citizens’ place of residence (so 0-500 meter compared to 500-1500 meter or 1500-4500 meter) reduces citizens' intended and revealed voting in a direct democratic in favor of renewable energies (the so-called Electricity law) compared to citizens’ provided with no such information or with information about a small Agri-PV developing potential in citizens’ close proximity.
H4b: Information about the realistic potential (small or large) for developing Agri-PV projects in the proximity to citizens’ place of residence (i.e., 0-500 meter, 500-1500 meter, and 1500-4500 meter around individuals’ homes) affects the gap between citizens' intended (before the vote) and revealed (after the voted) voting behavior in a direct democratic vote on renewable energies (the so-called Electricity law).
H5: The characteristics of a planned Agri-PV project (i.e., Agri-PV type, size, ownership, distance to place of residence, impact on communities’ own production with food, impact on communities’ own production with energy, and impact on farmer income) affect citizens’ Agri-PV project support.
Experimental design
Experimentally varied information treatment
During the first survey wave, participants will be randomly assigned to either the control (⅓ of all participants in the sample) or the treatment group (⅔ of participants in the sample). The control group will get some generic information on the potential of developing Agri-PV installations in Switzerland. They are informed that only part of the suitable land actually needs to be developed. The treatment group gets the same information and additionally is provided with a map that shows the surroundings of their place of residence. Indicated on the map is the realistic potential for building Agri-PV systems around their place of residence. The potential is identified either as high or low within a radius of 0-500m, 500-1500m and 1500-4500m around individuals’ homes. The data shown to individuals in the treatment group is based on a recent study by ZHAW, accounting for the realistic building and solar power generation potential of different Agri-PV installation types in the whole of Switzerland (Anderegg et al., 2024).
In a comprehension and manipulation check section of the survey, respondents in both the control and the treatment groups are asked several questions regarding the potential of Agri-PV installations in the proximity their home to make sure they understood the content of the treatment and investigate potential differences between the groups in terms of knowledge and concern about building Agri-PV plants close to respondents’ home. The map containing the Agri-PV potential is then presented again to the treatment group to foster the treatment effect.
Due to randomization, the only systematic difference between the control and the treatment group should be the treatment, which allows estimating a causal effect of the treatment on the dependent variables, if the randomization works as intended and the sample is generally large enough (Stock and Watson 2020).
Conjoint Experiment
We also measure support for differently designed Agri-PV systems in a conjoint experiment. The following paragraphs outline the introduction to the experiment, the seven policy package attributes, and the randomly varied policy attribute levels (see further details about the conjoint design and its randomization in the section Experimental Design and Randomization Method below).
We introduce the conjoint experiment to respondents as follows:
The next part of the survey will now look at possible concrete projects for the expansion of solar energy in Switzerland. In particular, we are interested in your opinion on agri-photovoltaics, a dual land use concept that combines the cultivation of agricultural crops with the installation of photovoltaic systems on land. Specifically, we will conduct a thought experiment with you in which we ask you to compare and evaluate alternative projects for the expansion of agri-photovoltaics in Switzerland. Each project has different aspects that can influence your evaluation.
Below you will find the most important aspects and characteristics that distinguish possible Agri-PV systems in Switzerland.
-Type of Agri-Pv system
-Size of the Agri-PV system
-Distance of the Agri-PV system from your own place of residence
-Ownership structure of the Agri-PV system
-Effect of the Agri-PV system on the crop yield of the agricultural area
-Effect of the Agri-PV system on the income of the farmers
After this introduction to the thought-experiment, in four different choice rounds, we then present participants with two alternative Agri-PV project options side by side. In each round, respondents compare the two options for Agri-PV systems carefully and indicate which option they prefer and to what extent they support or reject the respective project.
The different randomized Agri-PV project attributes and levels are the following:
Type of Agri-Pv system
- Agri-PV systems on greenhouses and replacement of polytunnels
- Vertical open space Agri-PV systems on pasture or arable land
- Horizontal open space Agri-PV system on pasture or arable land
Size of the Agri-PV system
- Up to one football pitch (approx. 1ha)
- Up to 5 football pitches (approx. 5ha)
- Up to 10 football pitches (approx. 10ha)
Distance of the Agri-PV system from your own place of residence
- 0-500 meters
- 500-1500 meters
- 1500-4500 meters
Ownership structure of the Agri-PV system
- Energy cooperative (e.g. local population)
- Farmers
- Landowner
- Municipality
- Regional energy supplier
- External (non-local) investors
Effect of the Agri-PV system on the crop yield of the agricultural area
- 0-5% reduction in crop yield
- 6-10% reduction in crop yield
- 11-20% reduction in crop yield
- 21-40% reduction in crop yield
- 41-80% reduction in crop yield
Effect of the Agri-PV system on the electricity production of the municipality
- 0-5% increase in own production
- 6-10% increase in own production
- 11-20% increase in own production
- 21-40% increase in own production
- 41-80% increase in own production
Effect of the Agri-PV system on the income of the farmers
- 0-5% higher income
- 6-10% higher income
- 11-20% higher income
- 21-40% higher income
- 41-80% higher income
Design details
Cited by
No citation data available for this registration yet.