![Sample our Environment & Agriculture journals, sign in here to start your access, latest two full volumes FREE to you for 14 days]({"type":"image" , "src" : "/sda/5934/TOC-Subject-Sample-2021-Environment-Agriculture.jpg"})
Abstract
This paper applies a choice experiment to estimate how Austrian households value the multiple impacts associated with future hydropower investments. Clear evidence was obtained that Austrian households are willing to pay for an expansion of hydropower capacities. More precisely, Austrians are willing to incur extra costs for economic (employment) and environmental (reduced air emissions) benefits related to the generation of electricity from hydroelectric power, but wish to be compensated for the loss of nature and landscape new hydropower plants are associated with. This relation can formally be denoted as the ‘energy–water trade-off’ imposed by the use of hydropower. Moreover, confirmation of the ‘Not in my backyard’ phenomenon was found, meaning that people prefer the construction of new hydropower stations, but not close to their homes.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Within the Kyoto Protocol Austria was bound to reduce its GHG emissions by 13% compared to the base year 1990 over the five-year period 2008–2012 (EU goal: −8%). Agreement over the second commitment period 2013–2020 has been reached at the climate conference in Doha 2012. Hence, a new global climate agreement should be concluded by the end of 2015 and is planned to come into force in 2020 (Umweltbundesamt Citation2013; UNFCC Citation2008).
2. Run-of-river plants transform the power of the water flowing in a river into electric energy, delivering mainly base-load power. The amount of producible electricity depends on the penstock and the natural water flow. Run-of-river schemes have little or no storage possibilities. The most important run-of-river plants are located in the eastern part of Austria along the major rivers like Danube, Inn, Enns, Mur and Drave. Storage schemes, by contrast, have water reservoirs to store the water. Hence, the generation of electricity can be decoupled from the timing of rainfall or glacial and snow melt. The capacity of a storage power plant depends on the altitude difference between the water reservoir and the power house. In case of high power demand the plants can immediately start running, i.e. electricity can exactly be generated when it is required. Hence, storage power plants are usually running at peak times (peak load). A special form of storage plants are pumped storage hydropower technologies. Here, off-peak electricity is used to pump water from a river or lower reservoir up to a higher reservoir to allow its release during peak times. The large storage power plants are exclusively located in the alpine areas of western Austria (Irena Citation2012; Kleinwasserkraft Österreich Citation2012).
3. This value corresponds to the estimate of reduced techno-economic potential which excludes potentials located in regions with a high degree of sensibility such as national parks and world heritages. However, the indicated hydropower potential does not consider reductions due to the possible restrictions imposed by the European WFD.
4. This target value includes the effect of revitalisation measures and the extension of existing facilities.
5. Besides that, some large-scale pumped storage power plants are especially planned to be built in order to balance the fluctuating electricity supply from wind power and photovoltaics.
6. The main objective of the WFD is to achieve a good ecological and chemical status for ground and surface water bodies until the year 2015. For the category of ‘heavily modified water bodies’ the aim of a good ecological potential is applied. For the determination of the water body status, biological, hydromorphological and physico-chemical quality elements are considered. Additionally, the WFD represents a legal framework to ensure that human agency as for instance the construction of new hydropower plants does not lead to a deterioration of the water body status (EPC Citation2000; BMLFUW Citation2006; BMLFUW Citation2007).
7. The concept of non-use values refers to the value that people derive from a good or service independent of any (present or future) use (Hausman Citation1993; Liebe and Meyerhoff Citation2005).
8. At EU level 410,000 net additional jobs in the renewable energy sector are assumed to be created by the adoption of renewable energy support policies that lead to an achievement of the 20% renewable energy goal by 2020 (Ragwitz et al. Citation2009).
9. For instance, the inability of fish to migrate up the river due to the existence of dams, alteration of flow conditions or reduced water levels downstream of the hydropower plant.
10. An analysis of Austrian waters revealed that the bulk of the rivers show significant hydromorphological impacts, mainly caused by hydroelectric use (Stigler et al. Citation2005; BMLFUW Citation2010b).
11. On the one hand, this selection of the CE attributes is based on an extensive literature review of stated preference applications on renewable energy and hydropower using landscape and environmental impacts, employment effects and reductions in air pollution as the relevant attributes (Bergmann, Colombo, and Hanley Citation2008; Longo, Markandya, and Petrucci Citation2008; Kataria Citation2009; Ku and Yoo Citation2010). On the other hand, the ‘right’ choice of attributes was addressed in several discussion rounds with external experts, as well as in the pre-test arriving at the final set of attributes explained above.
12. More precisely, three pilot studies using face-to-face interviews, postal and online surveys were conducted. Within the scope of the first pre-test round about 40 respondents received the draft questionnaire on hydropower expansion. Based on the results of the first pre-test, the (revised) questionnaire was retested on a sample of 55 people. The last pilot survey was conducted by a professional survey agency using a sample of 290 respondents. Based on this ‘test–retest’ procedure a well-founded questionnaire was developed able to be used in the final survey.
13. The group of experts (‘Advisory Board’) consisted of people from the Austrian electricity industry and different federal and provincial government departments.
14. Due to the fact that the implementation of the survey was outsourced to a professional market research institute, we neither have information on the time span needed to fill out the questionnaire nor on questionnaire exhaustion. However, the market research institute provides incentives for their panel participants. In particular, the survey participants receive points for filling out the questionnaire; the collected points can then be exchanged for real money.
15. In total, there were 73 respondents in the sample choosing the opt-out in each of the six choice sets. As mentioned above, only 13 out of these observations were categorised as ‘true’ protests. The identified protest voters are – as compared to the total sample – older, lower educated with less income, very poorly informed on hydropower and finally, substantially more negative towards the construction of new hydropower plants along Austrian rivers.
16. For more information see http://www.marketagent.com.
17. Vienna was included simply due to the fact that it represents the capital of Austria and therefore cannot be ignored when conducting a nationwide survey.
18. The underrepresentation of older people may be due to the data collection method, since the older population is usually less familiar with online surveys or the internet in general.
19. This represents an ambiguous result since the higher educated sample should usually result in a disproportional mean or median income. Hence, the individual answers to the income question in the survey may probably not be reliable.
20. The mean distance amounts to 27.9 km with a standard deviation of 53.5 km. This disparity between mean and median is indicative for the presence of outliers. Generally, outliers are defined as observations lying above the 75-percentile plus 1.5 times the interquartile range (Quatember Citation2014). Accordingly, in our sample outliers are observations with a stated distance of more than 67.5 km. This applies to 99 out of the 892 observations representing a quite high share of outliers.
21. The most widely used test for violation of the IIA assumption is the so-called Hausman test, developed by Hausman and McFadden in 1984 (Louviere, Hensher, and Swait Citation2000).
22. In the classical MNL model, each parameter in the indirect utility specification Vin is assumed to be a fixed estimate, i.e. equal for each individual (Hensher, Rose, and Greene Citation2005).
23. The panel data structure was accounted for in all of the estimated models.
24. Cardinal-linear coded variables were assumed to be normally distributed, while dummy coded attributes were specified to be uniformly distributed. More on random parameter distributions can be found in Hensher and Greene (Citation2002).
25. Appreciation of renewable energy sources is measured by respondents’ answer to the question whether it is important for them that their consumed electricity stems from renewable sources like hydropower, wind power or solar energy (see Table A1).
26. This result is also reflected by the general statistical analysis. Hence, there is a significant relationship (Cramer's V = 0.148, p-value = 0.000) between the level of information and respondents’ attitude towards the expansion of hydroelectric power in Austria. Correspondingly, poorly informed people have a more reluctant attitude towards hydropower expansion than better informed individuals.
27. The educational level may represent a proxy variable for income, since higher education is normally associated with higher incomes. And income is usually a strong determinant of stated choices.
28. More precisely, a negative MWTP can be interpreted as a reduction of individual utility resulting from the environmental changes (Meyerhoff, Ohl, and Hartje Citation2008).
29. According to Statistik Austria (Citation2011), the number of households in the area of investigation is 2,257,000.
30. However, it must be kept in mind that the sample is slightly higher educated as compared to the Austrian population and education usually represents a positive determinant of stated choices. Due to this fact estimates of MWTP, EWF and aggregated EWF may be biased upwards. In order to address this problem, a sensitivity analysis resulting in a range of (aggregated) EWF may increase the robustness of the results.