TD
T.W.T. Dieltjes
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
1 records found
1
The Effects of Wind Turbines on Housing Prices in the Netherlands
Assessing Onshore, Offshore, and Heterogeneous Effects
The Netherlands is rapidly expanding its wind energy capacity both onshore and offshore as part of its broader energy transition. Although wind capacity has more than doubled from 2019 to 2024, residents experience localised externalities from wind turbines, including visual impact, noise, and shadow flicker. These costs are often capitalised into housing prices. Therefore, the placement of turbines involves balancing the progress of the energy transition with the burdens imposed on local communities, emphasising the importance of reliable evidence on the extent and distribution of these effects to inform spatial planning strategies.
Among the many externalities of wind turbines, ranging from the effects on landscape to impacts on nature, housing prices are particularly relevant. They can be directly observed in housing transaction records and combine various perceived local disamenities into a single, quantifiable measure. Furthermore, house prices are linked to Dutch compensation policy: the Omgevingswet, or the Dutch Environment and Planning Act, sets a societal risk threshold of four per cent of the property’s value. If the depreciation falls below this threshold, it is not eligible for any compensation. Therefore, analysing house prices provides a tangible, policy-relevant insight into who bears the local costs of the energy transition. This aligns with the Engineering and Policy Analysis (EPA) programme’s goal of employing quantitative methods to inform decision-making regarding grand societal challenges.
While prior studies have examined how turbines impact housing prices in the Netherlands, two important gaps still persist. First, the impact of offshore turbines on housing prices has not been examined in the Dutch context. Second, existing studies capture spatial variation through geographic categories but leave the role of socioeconomic factors almost entirely unexplored. Additionally, the Dutch estimates are based on methods that could introduce bias as turbine adoption is staggered. This raises questions about the current magnitude of these effects when staggered adoption is taken into account, and how these effects are distributed among communities exposed to turbines.
This thesis estimates the effects of onshore wind turbines using a Difference-in-Differences (DiD) design, based on a sample of 1:83 million housing transactions from 2010 to 2023 from the Nederlandse Vereniging van Makelaars (NVM), merged with WindStats turbine data and neighbourhood-level Socioeconomic Status (SES) data. To extend current literature and to account for the staggered nature of Dutch wind turbine construction, the analysis uses the Sun & Abraham (2021) estimator and systematically controls for socioeconomic composition, which prior Dutch studies did not. A new addition to current literature is the offshore component, which is deliberately descriptive: It documents coastal price trends associated with offshore wind farm development and clarifies why the Dutch data do not support credible causal identification.
The results show that the effects of onshore wind turbines are small, not robust across different control group definitions, highly localised, and statistically indistinguishable from zero under most alternatives. In the preferred specification, they are found to slightly lower nearby property prices by about 0:8% within two kilometres. This is roughly a fifth of the four per cent compensation threshold of the Omgevingswet, so most affected homeowners do not qualify for any compensation. The estimated price effect is smaller than previous Dutch estimates, primarily because the sample from 2010 to 2023 limits the identifying variation needed to estimate causal effects. The causal interpretation is further limited by imperfect pre-trends, though robustness checks suggest these do not drive the estimated effect. These effects are unlikely to be lower due to changes in perceptions of wind energy, as Dröes & Koster (2019) have shown that willingness to pay to live next to turbines is generally stable over time.
The most robustly tested finding in the heterogeneity analysis is that additional turbines do not produce larger effects, meaning they do not lead to higher price effects, while also reducing the number of properties exposed to turbines. Subgroup analyses indicate effects are strongest near turbines and in densely populated regions, although these differences between subgroups are not statistically significant. Additionally, while turbines are disproportionately sited in neighbourhoods with lower transaction prices, associated with lower SES, whether the price effect itself differs by SES could not be established. In all these analyses, none of the estimated effects approaches the four per cent compensation threshold set by the Omgevingswet. The offshore wind analysis remains descriptive rather than causal due to the small number of offshore parks, their recent clustered construction, a limited coastal sample, and the lack of visibility data, all of which prevent reliable causal identification.
These findings inform a siting strategy focused on reducing housing-market costs of turbines. The only well-supported principle is that clustering turbines concentrates price effects on fewer properties, without increasing their magnitude relative to single-turbine installations. The weaker, non-significant patterns suggest, cautiously, that turbines should be placed away from urbanised areas. Since onshore turbines are already mostly located in communities with lower prices, associated with lower SES, an efficiency-based siting strategy that ignores distributional risks will exacerbate existing inequalities in turbine placement. Where turbines are sited is therefore ultimately a political choice that should weigh minimising housing market impacts against the fair distribution of turbine placements; this thesis explicitly highlights this trade-off by analysing both the magnitude of effects and the populations most affected. More broadly, the small estimated effects imply that turbines have a limited impact on the Dutch housing market, inviting a shift in the debate toward more difficult-to-measure externalities, such as landscape and health concerns. ...
Among the many externalities of wind turbines, ranging from the effects on landscape to impacts on nature, housing prices are particularly relevant. They can be directly observed in housing transaction records and combine various perceived local disamenities into a single, quantifiable measure. Furthermore, house prices are linked to Dutch compensation policy: the Omgevingswet, or the Dutch Environment and Planning Act, sets a societal risk threshold of four per cent of the property’s value. If the depreciation falls below this threshold, it is not eligible for any compensation. Therefore, analysing house prices provides a tangible, policy-relevant insight into who bears the local costs of the energy transition. This aligns with the Engineering and Policy Analysis (EPA) programme’s goal of employing quantitative methods to inform decision-making regarding grand societal challenges.
While prior studies have examined how turbines impact housing prices in the Netherlands, two important gaps still persist. First, the impact of offshore turbines on housing prices has not been examined in the Dutch context. Second, existing studies capture spatial variation through geographic categories but leave the role of socioeconomic factors almost entirely unexplored. Additionally, the Dutch estimates are based on methods that could introduce bias as turbine adoption is staggered. This raises questions about the current magnitude of these effects when staggered adoption is taken into account, and how these effects are distributed among communities exposed to turbines.
This thesis estimates the effects of onshore wind turbines using a Difference-in-Differences (DiD) design, based on a sample of 1:83 million housing transactions from 2010 to 2023 from the Nederlandse Vereniging van Makelaars (NVM), merged with WindStats turbine data and neighbourhood-level Socioeconomic Status (SES) data. To extend current literature and to account for the staggered nature of Dutch wind turbine construction, the analysis uses the Sun & Abraham (2021) estimator and systematically controls for socioeconomic composition, which prior Dutch studies did not. A new addition to current literature is the offshore component, which is deliberately descriptive: It documents coastal price trends associated with offshore wind farm development and clarifies why the Dutch data do not support credible causal identification.
The results show that the effects of onshore wind turbines are small, not robust across different control group definitions, highly localised, and statistically indistinguishable from zero under most alternatives. In the preferred specification, they are found to slightly lower nearby property prices by about 0:8% within two kilometres. This is roughly a fifth of the four per cent compensation threshold of the Omgevingswet, so most affected homeowners do not qualify for any compensation. The estimated price effect is smaller than previous Dutch estimates, primarily because the sample from 2010 to 2023 limits the identifying variation needed to estimate causal effects. The causal interpretation is further limited by imperfect pre-trends, though robustness checks suggest these do not drive the estimated effect. These effects are unlikely to be lower due to changes in perceptions of wind energy, as Dröes & Koster (2019) have shown that willingness to pay to live next to turbines is generally stable over time.
The most robustly tested finding in the heterogeneity analysis is that additional turbines do not produce larger effects, meaning they do not lead to higher price effects, while also reducing the number of properties exposed to turbines. Subgroup analyses indicate effects are strongest near turbines and in densely populated regions, although these differences between subgroups are not statistically significant. Additionally, while turbines are disproportionately sited in neighbourhoods with lower transaction prices, associated with lower SES, whether the price effect itself differs by SES could not be established. In all these analyses, none of the estimated effects approaches the four per cent compensation threshold set by the Omgevingswet. The offshore wind analysis remains descriptive rather than causal due to the small number of offshore parks, their recent clustered construction, a limited coastal sample, and the lack of visibility data, all of which prevent reliable causal identification.
These findings inform a siting strategy focused on reducing housing-market costs of turbines. The only well-supported principle is that clustering turbines concentrates price effects on fewer properties, without increasing their magnitude relative to single-turbine installations. The weaker, non-significant patterns suggest, cautiously, that turbines should be placed away from urbanised areas. Since onshore turbines are already mostly located in communities with lower prices, associated with lower SES, an efficiency-based siting strategy that ignores distributional risks will exacerbate existing inequalities in turbine placement. Where turbines are sited is therefore ultimately a political choice that should weigh minimising housing market impacts against the fair distribution of turbine placements; this thesis explicitly highlights this trade-off by analysing both the magnitude of effects and the populations most affected. More broadly, the small estimated effects imply that turbines have a limited impact on the Dutch housing market, inviting a shift in the debate toward more difficult-to-measure externalities, such as landscape and health concerns. ...
The Netherlands is rapidly expanding its wind energy capacity both onshore and offshore as part of its broader energy transition. Although wind capacity has more than doubled from 2019 to 2024, residents experience localised externalities from wind turbines, including visual impact, noise, and shadow flicker. These costs are often capitalised into housing prices. Therefore, the placement of turbines involves balancing the progress of the energy transition with the burdens imposed on local communities, emphasising the importance of reliable evidence on the extent and distribution of these effects to inform spatial planning strategies.
Among the many externalities of wind turbines, ranging from the effects on landscape to impacts on nature, housing prices are particularly relevant. They can be directly observed in housing transaction records and combine various perceived local disamenities into a single, quantifiable measure. Furthermore, house prices are linked to Dutch compensation policy: the Omgevingswet, or the Dutch Environment and Planning Act, sets a societal risk threshold of four per cent of the property’s value. If the depreciation falls below this threshold, it is not eligible for any compensation. Therefore, analysing house prices provides a tangible, policy-relevant insight into who bears the local costs of the energy transition. This aligns with the Engineering and Policy Analysis (EPA) programme’s goal of employing quantitative methods to inform decision-making regarding grand societal challenges.
While prior studies have examined how turbines impact housing prices in the Netherlands, two important gaps still persist. First, the impact of offshore turbines on housing prices has not been examined in the Dutch context. Second, existing studies capture spatial variation through geographic categories but leave the role of socioeconomic factors almost entirely unexplored. Additionally, the Dutch estimates are based on methods that could introduce bias as turbine adoption is staggered. This raises questions about the current magnitude of these effects when staggered adoption is taken into account, and how these effects are distributed among communities exposed to turbines.
This thesis estimates the effects of onshore wind turbines using a Difference-in-Differences (DiD) design, based on a sample of 1:83 million housing transactions from 2010 to 2023 from the Nederlandse Vereniging van Makelaars (NVM), merged with WindStats turbine data and neighbourhood-level Socioeconomic Status (SES) data. To extend current literature and to account for the staggered nature of Dutch wind turbine construction, the analysis uses the Sun & Abraham (2021) estimator and systematically controls for socioeconomic composition, which prior Dutch studies did not. A new addition to current literature is the offshore component, which is deliberately descriptive: It documents coastal price trends associated with offshore wind farm development and clarifies why the Dutch data do not support credible causal identification.
The results show that the effects of onshore wind turbines are small, not robust across different control group definitions, highly localised, and statistically indistinguishable from zero under most alternatives. In the preferred specification, they are found to slightly lower nearby property prices by about 0:8% within two kilometres. This is roughly a fifth of the four per cent compensation threshold of the Omgevingswet, so most affected homeowners do not qualify for any compensation. The estimated price effect is smaller than previous Dutch estimates, primarily because the sample from 2010 to 2023 limits the identifying variation needed to estimate causal effects. The causal interpretation is further limited by imperfect pre-trends, though robustness checks suggest these do not drive the estimated effect. These effects are unlikely to be lower due to changes in perceptions of wind energy, as Dröes & Koster (2019) have shown that willingness to pay to live next to turbines is generally stable over time.
The most robustly tested finding in the heterogeneity analysis is that additional turbines do not produce larger effects, meaning they do not lead to higher price effects, while also reducing the number of properties exposed to turbines. Subgroup analyses indicate effects are strongest near turbines and in densely populated regions, although these differences between subgroups are not statistically significant. Additionally, while turbines are disproportionately sited in neighbourhoods with lower transaction prices, associated with lower SES, whether the price effect itself differs by SES could not be established. In all these analyses, none of the estimated effects approaches the four per cent compensation threshold set by the Omgevingswet. The offshore wind analysis remains descriptive rather than causal due to the small number of offshore parks, their recent clustered construction, a limited coastal sample, and the lack of visibility data, all of which prevent reliable causal identification.
These findings inform a siting strategy focused on reducing housing-market costs of turbines. The only well-supported principle is that clustering turbines concentrates price effects on fewer properties, without increasing their magnitude relative to single-turbine installations. The weaker, non-significant patterns suggest, cautiously, that turbines should be placed away from urbanised areas. Since onshore turbines are already mostly located in communities with lower prices, associated with lower SES, an efficiency-based siting strategy that ignores distributional risks will exacerbate existing inequalities in turbine placement. Where turbines are sited is therefore ultimately a political choice that should weigh minimising housing market impacts against the fair distribution of turbine placements; this thesis explicitly highlights this trade-off by analysing both the magnitude of effects and the populations most affected. More broadly, the small estimated effects imply that turbines have a limited impact on the Dutch housing market, inviting a shift in the debate toward more difficult-to-measure externalities, such as landscape and health concerns.
Among the many externalities of wind turbines, ranging from the effects on landscape to impacts on nature, housing prices are particularly relevant. They can be directly observed in housing transaction records and combine various perceived local disamenities into a single, quantifiable measure. Furthermore, house prices are linked to Dutch compensation policy: the Omgevingswet, or the Dutch Environment and Planning Act, sets a societal risk threshold of four per cent of the property’s value. If the depreciation falls below this threshold, it is not eligible for any compensation. Therefore, analysing house prices provides a tangible, policy-relevant insight into who bears the local costs of the energy transition. This aligns with the Engineering and Policy Analysis (EPA) programme’s goal of employing quantitative methods to inform decision-making regarding grand societal challenges.
While prior studies have examined how turbines impact housing prices in the Netherlands, two important gaps still persist. First, the impact of offshore turbines on housing prices has not been examined in the Dutch context. Second, existing studies capture spatial variation through geographic categories but leave the role of socioeconomic factors almost entirely unexplored. Additionally, the Dutch estimates are based on methods that could introduce bias as turbine adoption is staggered. This raises questions about the current magnitude of these effects when staggered adoption is taken into account, and how these effects are distributed among communities exposed to turbines.
This thesis estimates the effects of onshore wind turbines using a Difference-in-Differences (DiD) design, based on a sample of 1:83 million housing transactions from 2010 to 2023 from the Nederlandse Vereniging van Makelaars (NVM), merged with WindStats turbine data and neighbourhood-level Socioeconomic Status (SES) data. To extend current literature and to account for the staggered nature of Dutch wind turbine construction, the analysis uses the Sun & Abraham (2021) estimator and systematically controls for socioeconomic composition, which prior Dutch studies did not. A new addition to current literature is the offshore component, which is deliberately descriptive: It documents coastal price trends associated with offshore wind farm development and clarifies why the Dutch data do not support credible causal identification.
The results show that the effects of onshore wind turbines are small, not robust across different control group definitions, highly localised, and statistically indistinguishable from zero under most alternatives. In the preferred specification, they are found to slightly lower nearby property prices by about 0:8% within two kilometres. This is roughly a fifth of the four per cent compensation threshold of the Omgevingswet, so most affected homeowners do not qualify for any compensation. The estimated price effect is smaller than previous Dutch estimates, primarily because the sample from 2010 to 2023 limits the identifying variation needed to estimate causal effects. The causal interpretation is further limited by imperfect pre-trends, though robustness checks suggest these do not drive the estimated effect. These effects are unlikely to be lower due to changes in perceptions of wind energy, as Dröes & Koster (2019) have shown that willingness to pay to live next to turbines is generally stable over time.
The most robustly tested finding in the heterogeneity analysis is that additional turbines do not produce larger effects, meaning they do not lead to higher price effects, while also reducing the number of properties exposed to turbines. Subgroup analyses indicate effects are strongest near turbines and in densely populated regions, although these differences between subgroups are not statistically significant. Additionally, while turbines are disproportionately sited in neighbourhoods with lower transaction prices, associated with lower SES, whether the price effect itself differs by SES could not be established. In all these analyses, none of the estimated effects approaches the four per cent compensation threshold set by the Omgevingswet. The offshore wind analysis remains descriptive rather than causal due to the small number of offshore parks, their recent clustered construction, a limited coastal sample, and the lack of visibility data, all of which prevent reliable causal identification.
These findings inform a siting strategy focused on reducing housing-market costs of turbines. The only well-supported principle is that clustering turbines concentrates price effects on fewer properties, without increasing their magnitude relative to single-turbine installations. The weaker, non-significant patterns suggest, cautiously, that turbines should be placed away from urbanised areas. Since onshore turbines are already mostly located in communities with lower prices, associated with lower SES, an efficiency-based siting strategy that ignores distributional risks will exacerbate existing inequalities in turbine placement. Where turbines are sited is therefore ultimately a political choice that should weigh minimising housing market impacts against the fair distribution of turbine placements; this thesis explicitly highlights this trade-off by analysing both the magnitude of effects and the populations most affected. More broadly, the small estimated effects imply that turbines have a limited impact on the Dutch housing market, inviting a shift in the debate toward more difficult-to-measure externalities, such as landscape and health concerns.