Estimating Civic Awareness and Willingness to Pay (WTP) for Rooftop Greenhouses: A Study of Seoul Residents
Article information
Abstract
Background and objective
In hardscape-dominated cities, rooftop greenhouses(RGs) are considered a viable alternative for providing high-quality food, greenery, and experiences to citizens. However, existing research has predominantly focused on conventional urban agriculture, and rooftop greenhouses remain under-researched on an international scale. In this study, we use the contingent valuation method (CVM) to estimate the willingness to pay (WTP) for RGs among residents of Seoul, South Korea, and evaluate their comprehensive value.
Methods
A questionnaire including items on respondents’ characteristics, WTP, and civic awareness of RGs was prepared, and WTP was estimated using the double-bounded dichotomous choice (DBDC) model. On the basis of these copies, utility theoretic analysis and the logit model of CVM were employed to estimate the WTP for RG among Seoul residents.
Results
The per capita WTP for RG among Seoul residents was 30,260. On the basis of this result, the value of RGs in Seoul was approximately 220 billion KRW. Most respondents preferred a combination of crop production and experiences. This outcome suggests that future RGs in buildings should include spaces for citizen experiences and education. Respondents suggested that the most necessary measure to promote RGs was providing customized education on building and managing rooftop greenhouses, followed by providing and subsidies, and promoting related policies and projects.
Conclusion
The findings of this study can be utilized to convey the global importance of RGs and as a basis for establishing related policies and increasing the participation of local residents. In addition, considering that Seoul has recently implemented a project to install RGs on existing commercial buildings, the results of this study will be highly useful.
Introduction
Recent challenges such as shifts in cropping systems due to climate change, the emergence of novel infectious diseases, and food security crises stemming from wars have led to a projected 1.3–2.1% decrease in global food production in 2022/23 compared to 2020/21 (Lee, 2020; FAO, 2023). In response, the international community has increasingly focused on urban agriculture, producing crops in urban areas in which over half of the world's population resides (Veenhuizen, 2006), in order to increase food self-sufficiency rates. Urban agriculture can be defined as the cultivation of plants or the raising of animals within cities for various purposes, including food production (Veenhuizen, 2006; Bon, 2010). Urban agriculture is practiced in a range of urban land uses, such as building rooftops, balconies, indoor areas, and unused outdoor spaces. However, urban agriculture still faces limitations, including a lack of available space in urban areas, lower yield per unit area compared to rural agriculture, and concerns over the safety of crops exposed to pollutants.
To resolve these issues, rooftop greenhouses, where crops are cultivated in greenhouses installed on urban building rooftops, have recently emerged as a key alternative (Specht et al., 2014; Appolloni et al., 2021). Rooftop greenhouses are a new type of urban agriculture that doesn't require additional land, reducing space constraints and enabling local food production and consumption. Moreover, it maintains consistent crop quality regardless of disasters, seasonal changes, or air pollution, and enables vertical farming so that it is possible to cultivate higher crop yields per unit area compared to conventional rooftop gardens. In fact, tomato yields per unit area from rooftop greenhouses in Barcelona, Spain, were 1.4–3.8 times higher than those from traditional gardens (Orsini et al., 2014; Grard et al., 2015; Harada et al., 2017; Sanjuan-Delmás et al., 2018; Boneta et al., 2019). Given these advantages such as stable crop production and reduced food miles, rooftop greenhouses are increasingly being commercialized in various developed countries (Appolloni et al., 2021).
Rooftop greenhouses are an emerging trend in the United States and Europe, with notable urban applications such as Whole Foods Market in the U.S. and Cité Maraichère in France. In Seoul, the capital city of South Korea, urban agriculture encompasses approximately 222 hectares as of 2020, representing a 7.7-fold increase since 2011 (Seoul, 2024). Despite the growing interest in urban agriculture, there have been no rooftop greenhouses in Seoul until now. However, the installation and operation of the first rooftop greenhouse at a commercial building in Seongsu-dong in late 2023 are anticipated to increase interest and demand. These do not simply stop at producing crops but also demonstrate positive functions in economic, environmental, and social aspects, such as reducing carbon footprints, saving heating and cooling energy for buildings, providing green experiences, creating multifunctional cultural spaces, generating jobs for local residents, and improving urban landscapes. Rooftop greenhouses, as mentioned above, provide various functions to urban residents, but as they are in the early stage of technology, the focus has been placed more on predicting investment costs and developing related technologies rather than comprehensively diagnosing their value. According to prior studies (Mandel, 2013; Proksch, 2016; Milford, 2019; Sanyé-Mengual et al., 2015), the investment cost per unit area of rooftop greenhouses ranged from 230 to 764 USD/m2.
Although there are no prior studies evaluating the value of rooftop greenhouses, numerous studies on related fields such as urban agriculture, rooftop greening, and urban green spaces have been conducted worldwide (Hong et al., 2005; Jang et al., 2006; Heo and Kwon, 2014; Kim et al., 2014; Atmaja et al., 2020; Neckel et al., 2020; Gustavsen et al., 2022). For urban agriculture, the willingness to pay (WTP) per household by residents in Seoul and Jeonju, South Korea, was 17,034–23,041 KRW and 31,476 KRW, respectively (Jang et al., 2006; Heo and Kwon, 2014). Residents in Oslo, Norway, and Malang, Indonesia, responded that they were willing to pay 15.8–21.6 EUR/m2 and 28.68 USD/m2, respectively, for urban agriculture (Atmaja et al., 2020; Gustavsen et al., 2022). According to Kim et al. (2014), the WTP per household for improving urban aesthetics, providing relaxation spaces, and enhancing comfort in Seoul’s rooftop greening projects was 17,222 KRW. Similarly, residents in Seoul, South Korea, and Passo Fundo, Brazil, responded that they were willing to pay 24,000 KRW and 30.68 USD, respectively, for the expansion and maintenance of urban green spaces (Hong et al., 2005; Neckel et al., 2020).
Most of the aforementioned studies tended to evaluate the value based on citizens’ WTP using the contingent valuation method (CVM). In order to enhance the utility of emerging technologies, such as rooftop greenhouses, it is essential to reflect the opinions and perceptions of actual users, the citizens. WTP is an effective tool for efficiently incorporating citizens' views in the process of identifying alternatives that serve the public interest (You and Kim, 1999). Methods for evaluating WTP include CVM, the travel cost method, choice experiment, hedonic price mode, and the avoidance expenditure method (Kang et al., 2011). Choice experiments are a method for indirectly estimating WTP by incorporating costs into the attributes of hypothetical alternatives (Boxall et al., 1996; Korea Development Institute, 2012). This approach is advantageous for capturing respondents' preferences in detail; however, it can be challenging to produce reliable results when respondents have a limited understanding of the attributes (Oh et al., 2020). Travel cost methods and hedonic price models depend on current data, including actual travel expenses and market transaction information, to estimate value. This reliance can restrict analysis when such data is unavailable (Oh et al., 2020). Among these, CVM is a method that surveys consumers' WTP by setting up a hypothetical market (Mitchell and Carson, 1989). It has the advantage of comprehensively evaluating public goods, such as improving urban landscape and enhancing resident health, in addition to direct values like crop sales. Unlike previous methods, this approach is particularly beneficial for intuitive valuation in situations where citizens have limited understanding and relevant tangible data is scarce, such as in the case of rooftop greenhouses.
Meanwhile, to enhance the utility and applicability of rooftop greenhouses, which are in the early stages of adoption, it is necessary to comprehensively evaluate the diverse values they provide, perceive their importance, and prepare foundational data to prove the validity of policy support. However, as mentioned earlier, there is insufficient research worldwide that evaluates the value of rooftop greenhouses. Thus, the purpose of this study is to investigate the WTP for rooftop greenhouses using CVM with Seoul residents as the target population, and based on this, analyze the comprehensive value and civic awareness of rooftop greenhouses. Given that Seoul has recently implemented a project to install rooftop greenhouses on existing commercial buildings, the findings of this study are expected to have high practical utility. Furthermore, the results of this study can be effectively used as foundational information for sharing the importance of rooftop greenhouses internationally, establishing related policies, or expanding the participation of local residents.
Research Methods
Questionnaire design and survey
This study designed the questionnaire based on the assumption that a rooftop greenhouse development project is being implemented in Seoul. The survey items were selected based on related prior studies (Hong et al., 2005; Jang et al., 2006; Heo and Kwon; 2014; Kim et al., 2014; Atmaja et al., 2020; Neckel et al., 2020; Gustavsen et al., 2022) and expert consultations, specifically including items such as WTP, preferred types and activities for rooftop greenhouses, intentions to purchase crops, and measures to promote rooftop greenhouses (Table 1). The specific advice requested from the experts was for a valuation method that efficiently assesses the environmental, economic, and social value of rooftop greenhouses in an integrated manner, as well as a questionnaire to gauge the perceptions of rooftop greenhouses among residents of Seoul. The method for presenting WTP was proposed as a fund, reflecting practical feasibility. Meanwhile, WTP surveys using CVM can be categorized into open-ended and closed-ended formats. Open-ended surveys involve directly entering the price, whereas closed-ended surveys present a specific amount and ask respondents to answer in a dichotomous choice (DC) of “yes” or “no” (Hanemann, 1984). Prior studies using CVM tended to adopt DC as the primary questioning method due to its ease of response and the absence of starting point bias (Lee, 2013; Kim and Park, 2023).
The DC questioning format is divided into single-bounded dichotomous choice (SBDC) and double-bounded dichotomous choice (DBDC), and their specific details are as follows. SBDC involves asking and responding to a proposed amount only once, while DBDC presents a second amount that is either twice as high or half as low as the first proposed amount, depending on the response to the first question, and asks about WTP again (Hanemann et al., 1999). This study utilized DBDC, which has a relatively higher response rate and can reduce the occurrence of irrational WTP, to design the WTP questionnaire.
Meanwhile, the DBDC applied in this study for estimating WTP has been found to exhibit underestimation bias (Carson et al., 1999) or a lack of consistency in responses between the first and second questions (Cameron and Quiggin, 1994; McFadden, 1994). Accordingly, prior studies (Bateman et al., 2001; Kim and Park, 2023) recommend structuring the questionnaire in a DBDC format but analyzing only the response to the first question to reduce the underestimation bias of DBDC. Thus, this study also utilized only the first response for WTP analysis.
A preliminary survey was conducted to determine the WTP amounts to be presented in the questionnaire. Approximately 50 participants were surveyed in an openended format regarding the amount of funding they could pay if a rooftop greenhouse development project were implemented in Seoul. Excluding the top and bottom 10% of the responses (Lee et al., 2014), the distribution of amounts ranged from approximately 5,000 KRW to 80,000 KRW. Given this result, the proposed WTP amounts for rooftop greenhouse development in this study were set at 5,000 KRW, 10,000 KRW, 20,000 KRW, 40,000 KRW, and 80,000 KRW. The survey was conducted online, with the preliminary survey carried out for one week in June 2023, and the main survey conducted over two weeks in November 2023. The survey targeted residents of Seoul aged 20–60 who were actively engaged in economic activities, and the total number of valid responses collected was 1,008. To eliminate bias in the survey results related to demographics such as neighborhood and gender, the sample was stratified by borough and gender demographics.
Analysis method
In this study, WTP was estimated using the utility theory of CVM and a logit model, and the specific process and details are as follows. DC CVM is based on utility theoretic analysis (Hanemann, 1984), and its analytical model can be expressed in the form of Equation (1). Here, U is the indirect utility function, r is a binary indicator variable, where a value of 1 indicates there is WTP for rooftop greenhouses, and 0 indicates there is no WTP. Y represents the respondent's income, S refers to individual attribute variables, and ∈γ denotes the probability error, which is a random variable with a mean of 0 and is independent with identical distribution (Kim, 2010; Kim and Park, 2023).
When a respondent chooses “yes,” it indicates that they can obtain greater utility by paying the proposed amount A compared to when they choose “no.” This can be expressed as follows:
Based on the above, the probability of a respondent choosing “yes” can be specifically presented as the following equation. Here, Fθ[ΔV] represents the difference between the probability of paying the fund and the probability of not paying, and F refers to the probability distribution function of θ. The probability distribution function is typically estimated using either a probit model or a logit model. The logit model is more commonly used because it is easier to estimate and has better fit compared to the probit model (Lee and Mjelde, 2007; Kim, 2010). Therefore, this study also used the logit model to estimate WTP.
The estimation method of WTP using the logit model can be divided into the mean, median, and truncated mean, depending on the range of the proposed amounts. The mean is calculated by integrating the proposed amount from zero to infinity, while the median includes up to the negative proposed amount. The truncated mean includes the proposed amount from zero to the maximum value, which is the most commonly used approach. In this study, the mean, median, and truncated mean of WTP were all estimated, but the truncated mean was presented as the final WTP. These WTP estimations were conducted using R, while
Results and Discussion
Demographic characteristics
The respondents’ place of residence was analyzed by dividing the areas into the Central Region, Dongbuk (“northeast”) Region, Dongnam (“southeast”) Region, Seobuk (“northwest”) Region, and Seonam (“southwest”) Region. Specifically, the Central Region includes Jongno-gu, Jung-gu, and Yongsan-gu; the Dongbuk (“northeast”) Region includes Seongdong-gu, Gwangjin-gu, Dongdaemun-gu, Jungnang-gu, Seongbuk-gu, Gangbuk-gu, Dobong-gu, and Nowon-gu; the Dongnam (“southeast”) Region includes Seocho-gu, Gangnam-gu, Songpa-gu, and Gangdong-gu; the Seobuk (“northwest”) Region includes Eunpyeong-gu, Seodaemun-gu, and Mapo-gu; and the Seonam (“southwest”) Region includes Gangseo-gu, Yangcheon-gu, Yeongdeungpo-gu, Guro-gu, Geumcheon-gu, Gwanak-gu, and Dongjak-gu. The respondents were generally concentrated in the Dongbuk (“northeast”) Region (31.4%) and Seonam (“southwest”) Region (31.0%). The duration of residence showed that 59.1% of respondents had lived in their current location for less than 10 years, followed by 21.0% for 10–20 years, 12.0% for 20–30 years, 5.5% for 30–40 years, and 2.4% for more than 40 years (Table 2). As for gender, 50.3% of the respondents were male and 49.7% were female. The age distribution was as follows: 31.0% were in their 40s, 27.3% in their 30s, 20.9% in their 50s, 13.3% in their 20s, and 7.5% in their 60s. Regarding marital status, 56.4% of respondents were married, and among the married individuals, 85.1% had children.
Willingness to pay (WTP)
The payment probability for respondents based on the proposed amounts was highest for 5,000 KRW at 57%, followed by 10,000 KRW at 54%, and 20,000 KRW at 37% (Table 3). In other words, the payment probability tended to decrease as the proposed amount increased. The results of deriving a logit model based on the proposed amounts are shown in Table 4. The WTP for rooftop greenhouses of Seoul residents was calculated based on the logit model, and the results showed a mean of 46,847 KRW, median of 8,519 KRW, and truncated mean of 30,260 KRW. Applying the population of economically active Seoul residents aged 20–60, which is 7,336,801, it was estimated that the value of rooftop greenhouses is approximately 220 billion KRW.
In similar studies, residents of Seoul and Oslo, Norway, reported they could pay 17,034–23,041 KRW per household and 15.8–21.6 EUR per person for urban agriculture, respectively (Heo and Kwon, 2014; Gustavsen et al., 2022). Considering the average household size of 2.2 members in Seoul (Seoul, 2024), the WTP for urban agriculture per person in Seoul could be inferred to be between 7,740–10,470 KRW per person. The WTP estimated in this study was similar to the results from Oslo, but 3–4 times higher compared to Seoul's WTP. Despite also being in Seoul, the WTP in this study is higher compared to prior research (Heo and Kwon, 2014), likely due to the fact that rooftop greenhouses can produce high-quality crops year-round without seasonal constraints unlike urban farms and have the potential to serve as landmarks in the city, which may have been reflected positively in respondents’ perceptions. A previous study (The Seoul Institute, 2012) examining the perceptions of Seoul residents regarding urban farming found that the primary purpose of urban farming is to produce high-quality, healthy food. Additionally, the lack of available space was identified as the main factor inhibiting urban farming. In the case of Oslo, Norway, the short growing season due to the cold and wet climate is likely the reason for the relatively high willingness to pay for rooftop
Civic awareness of rooftop greenhouses
The type of rooftop greenhouse that respondents preferred the most was crop production + processing + experience at 38.3%, followed by crop production + experience at 26.9%, crop production + processing at 21.2%, crop production at 12.9%, and others at 0.7%. In other words, respondents tended to prefer types where they could engage in various experiences and education in addition to crop production (Table 5). Meanwhile, married respondents with children tended to prefer crop production + experience, which seems to reflect the expectation of providing environmental education to their children (Fig. 1). The types of experiences and education preferred by respondents in rooftop greenhouses were led by crop harvesting experiences at 30.7%, followed by energy reduction and environmental education at 22.1%, crop processing and eating and drinking experiences at 21.4%, smart agriculture specialized education at 17.4%, performances, exhibitions, and events at 8.3%, and others at 0.1%.
For measures to promote rooftop greenhouses, the highest response rate was for providing customized education on building/managing rooftop greenhouses at 22.0%, followed by providing incentives and subsidies at 19.0%, promoting related policies and projects at 19.0%, developing and disseminating rooftop greenhouse business modules at 18.4%, and identifying high-value crops at 12.7% (Table 6). The respondents generally wanted to purchase crops from rooftop greenhouses (64%), and 70% of all respondents answered that they have the intention to visit rooftop greenhouses in Seoul once they are completed.
Conclusion
Amid global issues such as climate change and food crisis, there has been growing interest in rooftop greenhouses, which involve installing greenhouses on the rooftops of urban buildings to grow crops. In cities dominated by hardscapes, rooftop greenhouses play a crucial role in promoting urban sustainability by enhancing food self-sufficiency and offering green landscapes and green experiences for urban residents. To improve the effectiveness and utilization of rooftop greenhouses, it is essential to conduct a comprehensive evaluation of the various functions they provide and to present evidence of their significance. Accordingly, this study evaluated the WTP for the development of rooftop greenhouses using CVM as well as the overall value of these rooftop greenhouses targeting residents living in Seoul, the capital city of South Korea, and analyzed their civic attitudes, including preferred activities and types of experiences.
A survey was conducted in November 2023 with Seoul residents regarding their WTP and civic awareness for rooftop greenhouses, resulting in 1,008 valid responses. The data analysis revealed that the WTP per resident for the establishment of rooftop greenhouses was 30,260 KRW. When applying this to the economically active population aged 20–60, the overall value of rooftop greenhouses was estimated at approximately 220 billion KRW. Meanwhile, the WTP for rooftop greenhouses in this study showed trends similar to those found in a previous study on urban agriculture in Oslo, Norway.
Most respondents preferred rooftop greenhouses that combine crop production and experience, and this trend was particularly evident among married individuals. This is believed to reflect an expectation to provide environmental education for their children. In fact, the most preferred type of experience and education for rooftop greenhouses was crop harvesting experience (30.7%), followed by energy reduction and environmental education (22.1%), and crop processing and eating and drinking experiences (21.4%). Respondents suggested that the most necessary measure to promote rooftop greenhouses was providing customized education on building and managing rooftop greenhouses. This was followed by providing incentives and subsidies, promoting related policies and projects, and developing and disseminating rooftop greenhouse business modules. Furthermore, most respondents (70%) answered that they have an intention to visit rooftop greenhouses in Seoul once they are completed.
In summary, the results suggest that Seoul residents generally had positive perceptions toward rooftop greenhouses and preferred models that integrate crop production with experience and education. This implies that it is necessary to also consider spaces for citizen experiences and education when installing rooftop greenhouses on buildings. Moreover, to promote rooftop greenhouses, there is a need to develop guidelines for the establishment and operation of rooftop greenhouses that reflect various building types in urban areas, along with incentive systems to encourage their adoption and use. This study is limited by the fact that the WTP survey was conducted only among Seoul residents, and it did not capture their experiences with urban agriculture Future studies should include residents from other cities as well to calculate WTP on a national scale in light of regional differences, based on which they can build advanced foundational data to establish and implement relevant policies.
Notes
This work was supported by the Korea Institute of Energy Technology Evaluation and Planning (KETEP) and the Ministry of Trade, Industry & Energy (MOTIE) of the Republic of Korea (20212020800050).