How do You Want to restore?--Assessing the Public Preferences and Social Benefits of Ecological Restoration for Natural Rubber Plantation in China

Revealing the public’s preference for ecological restoration projects will help increase public support and improve social benefits evaluation accuracy, which is a prerequisite for implementing ecological restoration projects. This study aimed to reveal the public’s preference for natural rubber plantation restoration projects and then quantify the social benefits of these projects to provide valuable references for related policy design. Based on choice experiments, we built a hypothetical market of ecological products and conducted a field survey to obtain the public’s preferences and willingness to pay (WTP) for natural rubber plantation restoration projects. Then, a random parameters logit model was applied to obtain public preference information, and then the social benefits brought by different restoration scenarios were calculated and compared. The results showed that: 1) residents of Hainan Province had a positive attitude toward ecological restoration for natural rubber plantations and were willing to bear a personal cost. 2) respondents had significant differences in preference for attributes of restoration projects. 3) a restoration project covering 2.1 million mu of rubber plantations, prioritizing the water source protection area, focusing the implementation on state farms, and developing the under-forest breeding economy would gain more social benefits, precisely, 337.543 million yuan/year. Ecological restoration policies should pay more attention to public needs and incorporate them into future guidelines. Policymakers should focus on restoration quality instead of the restoration area. Priority should be given to rubber plantations close to the water source protection area and small householders, and an appropriate under-forest economy should be considered. This study can provide a valuable reference for policy-making related to rubber plantation restoration.

Introduction

The introduction of natural rubber cultivation in China’s tropical areas has triggered unprecedented economic development, particularly in rural areas. Rubber farming has contributed to the local economy, increased smallholders’ income, and reduced poverty (Ahlheim et al., 2015; Ma et al., 2019). As rubber plantations are generally quite lucrative and have provided considerable cash income to farmers who have few alternative sources of income (Min et al., 2019), rubber plantations are encroaching the indigenous rain forests at a large scale and a high speed in the past 2 decades (Warren-Thomas et al., 2015) Additionally, because rubber can be successfully grown in only a small portion of southern tropical China, the strong domestic demand for rubber combined with government policies has caused substantial natural forest loss in China’s most biodiverse areas (Yi et al., 2014). As a consequence of the rapid expansion of natural rubber farming, dramatic changes in land use and ecosystems in southern China have occurred, leading to severe environmental problems, such as ecological degradation, biodiversity loss, and soil erosion (Ahrends et al., 2015; Wigboldus et al., 2017; Xiao et al., 2019).

The negative ecological effects of monoculture rubber plantations have recently attracted scientific and governmental concern and led scientists to call for the restoration of natural forests in some geographic regions. Local governments have made the restoration and protection of ecosystems a major policy issue. However, most of China’s policy design in the past has been top-down, which is easy to ignore the actual needs of the policy audience, and the effect was often not ideal, sometimes even disastrous (Zhang et al., 2011). As a result, restoration plans of rubber plantations often neglect local stakeholders and the multi-scalar politics in which they are entwined, thereby jeopardizing social and ecological conditions. Admittedly, a decentralized and participatory governance model, which is often characterized as “bottom-up,” has been increasingly advocated to achieve environmental policy objectives in a more sustainable and effective manner, as it attends heavily to local interests and searches for context-adaptive solutions to local environmental problems (Ouyang et al., 2020). Facts have also proved that the “bottom-up” policy design is conducive to improving public support and implementation effects (Huang and Kim, 2020; Ouyang et al., 2020). Benefits of the “bottom-up” approach have been demonstrated in many studies, such as the practice of participatory forest management (PFM) and participatory agroforestry development for restoring degraded sloping, which empowered local communities or villagers to own, manage and use the forest resources. (Xu et al., 2012; Gashu and Aminu, 2019). With regard to rubber plantation restoration, it is also essential to fully understand and incorporate public preferences in the policy design.

Many scholars previously focused on the importance of ecological restoration projects and problems faced in rubber plantation restoration, while others have turned their attention to the issues of farmers’ compensation and participation in ecological restoration (Grist and Menz, 1995; Smajgl et al., 2015; Gan et al., 2021). The primary forest’s rapid transformation into a single rubber plantation helped improve the economic situation of local families. However, it also posed a threat to ecological diversity and the original forest landscape, the massive expansion of rubber hurt the ecological and economic system (Liu et al., 2013; Chen et al., 2016). Restoration projects of rubber plantations currently implemented in China aimed to establish a composite ecosystem, increase biodiversity, and improve ecosystem service capacity (Sturgeon et al., 2014; Bingli et al., 2020). By contrast, a relatively small number of families preferred to reduce their rubber planting areas to eliminate the negative impact on the ecology (Hammond et al., 2017; Widianingsih et al., 2019). Min investigated 612 rubber farmers’ willingness to participate in ecological protection in Xishuangbanna, and found that relatively affluent families tended to provide expenses or labor for ecological restoration (Min et al., 2018). As a result, rubber plantations restoration projects should be implemented based on a thorough understanding of public preferences and values.

Theoretically, ecological rubber plantations have strong public goods attributes. Because of its non-market characteristics, the social benefits (social welfare improvement) are often underestimated or ignored, leading to an insufficient supply (Delaney and Jacobson, 2014; Blanco et al., 2018; Blanco et al., 2021). Like most public goods, ecological restoration of rubber plantations faces the same supply dilemma. In the context of the national pilot zone for ecological conservation, the Chinese government has emphasized that agricultural development must be environmentally friendly and conducive to the conservation of ecological conditions. Hence the “Environment-Friendly Rubber Plantation” program, which was proposed in 2009, has been gradually implemented by local governments. Due to the efficient and sustainable future investments in restoration infrastructure and improved services, it is suggested that the government and management agencies should make reasonable payment plans to pay for the supply of ecosystem services (Smajgl et al., 2015). Decision-makers need to weigh the associated costs and benefits of different rubber plantation restoration plans. Therefore, the essential premise is to evaluate the social benefits of rubber plantation ecological restoration to serve as a foundation for relevant ecological governance policies.

In terms of the social benefits of rubber plantation ecological restoration, most existing research has analyzed it qualitatively (Wu et al., 2001; Yi et al., 2014; Wigboldus et al., 2017), calculation from a quantitative perspective has still rarely been conducted (Ahlheim et al., 2015). This gap is partly due to the following reasons, making the market price system and general evaluation methods challenging. On the one hand, rubber restoration projects act as public goods. The externality characteristic makes environmentally-friendly rubber plantations benefit a broader group who depend on the overall quality of the ecosystem. On the other hand, as rubber plantations provide various ecosystem services to the public, a monetary value under non-market valuation is not easily measured. Scholars recently have achieved fruitful results from evaluations of non-market values, which serve as a good reference for this study (Himes-Cornell et al., 2018; Hynes et al., 2021). The most frequently used non-market value evaluation methods include the CVM (contingent valuation method) and CE (choice experiments) (Jin et al., 2018; Yao et al., 2018). The central concept behind these two methods is to build a hypothetical market to simulate ecological product consumption behavior. By observing consumers’ behavior in hypothetical markets, researchers can reveal the impact of improved ecological conditions on respondents’ welfare and then calculate the non-market value of ecological products.

The difference between these two methods is that the CVM usually assesses the social welfare changes caused by ecological improvements from an overall perspective of ecological products (Johnston, 2007; Ruto and Garrod, 2008). There is only one ecological product in the hypothetical market, and it reflects the improvement in ecological conditions from the current state to a specific state. For this reason, CVM is considered to be more concise and easier to operate. By contrast, CE pays more attention to the multiple attributes of ecological products. It uses representative attributes to form products and derives various products through changes in attribute values. In other words, there are multiple ecological products in the hypothetical market constructed by the CE, and researchers can calculate the increased benefits brought by any possible state of the ecological condition (Yao et al., 2018; Xu et al., 2020). Thus, CE has greater flexibility and is closer to the real market trading scene. It also gives the respondents more opportunities to choose and weigh the pros and cons, thus revealing their preferences for different ecological restoration attributes (Johnston, 2007; Ruto and Garrod, 2008). Therefore, CE is considered the most promising research method and has been widely used to evaluate the social benefits of protection and restoration in many fields, such as watersheds, wetlands, forests, and grasslands (Schaafsma et al., 2012; Olsen et al., 2019; Cai et al., 2020; Petcharat et al., 2020).

Given the importance of rubber plantation restoration for the sustainable development of China, the current work aims to examine the preferences and WTP of residents in Hainan Province for rubber plantation restoration and estimate social benefits from the restoration projects by using a choice experiment approach. The evaluation results derived from different restoration projects may assist policymakers in deciding on which projects should be prioritized. So, we first used a CE to build a hypothetical market by simulating market trading behavior and then evaluated the public’s welfare improvement for different potentially possible restoration projects. In the specific analysis, we use the econometric model to reveal the difference in public preferences and then calculate the corresponding social benefits. We expect the findings can provide a scientific reference for the policy design of rubber plantation restoration and aid the design and implementation of rubber restoration projects.

Methods and Materials

Choice Experiments

The primary thoughts of CE are to create a hypothetical market with different ecological products and ask the preferences of respondents, which can inform us of their willingness to pay (WTP) for that specific ecological product. Ultimately, social benefits can be calculated by analyzing respondents’ ecological product consumption preferences with econometric models (Adamowicz et al., 1998). Like regular products, restoration projects can be regarded as ecological products consisting of specific restoration attributes with various prices. In the field survey, respondents are given a set of restoration projects (or “choice sets”) composed of several restoration attributes. The respondents can choose what they feel are the best ecological products by selecting the trade-offs between the different restoration attributes and prices. CE provides respondents alternative ecological products composed of multiple attributes at different levels compared with other evaluation methods, and we can obtain respondents’ utility changes from the attributes they select (Syuhada et al., 2020; Ureta et al., 2021). This method could examine how respondents make trade-offs between possible levels of rubber ecosystem services provided by restoring projects. Using responses to the choice experiments, we estimate respondents’ WTP. On this basis, the social benefits of different restoration projects can be calculated, thus providing more references for further policy design (Knickel and Maréchal, 2018; Ureta et al., 2021).

In accordance with procedures for choice experiments proposed by Hensher et al., we evaluated the social benefits of rubber plantation ecological restoration following the steps shown in Figure 1: 1) based on expert consultations, focus group interviews, and a pilot survey, we selected attributes at different levels which were described in a manner so that every respondent can understand. 2) multiple alternative restoration projects were generated by orthogonal experiments according to attributes we had developed, then various alternative projects were further combined into a choice set in which respondents were asked to choose their most preferred projects. 3) experimental sampling scale was determined according to the data collection requirements, and finally, the implied value of each restoration attribute and the social benefits brought by different restoration projects were calculated using an RPL (Random Parameter Logit)model based on the survey data.

FIGURE 1

FIGURE 1. Benefits evaluation based on CE.

Econometric Model

The theoretical basis of choice experiments is Lancaster’s multiattribute utility theory (Lancaster, 1966) and random utility theory (McFadden, 1973). According to them, the utility provided by a specific product can be decomposed into the sum of the utility provided by the attributes which compose it (Lancaster, 1966). Random utility theory assumes that individuals make utility maximization choices according to the trade-offs between different product features. Therefore, a CE transforms the selection problem into a utility comparison problem and uses the maximization of utility to represent the consumer’s choice of the optimal product. It is assumed that the utility U_ni obtained by respondent n when choosing i from Q alternative ecological products is as follows (Hensher and Greene, 2002; Heiss, 2016):

$U_{ni}=V_{ni}+{\epsilon }_{ni}(1)$

Where V_ni is the measurable portion of utility, representing the effect of observable factors on respondents’ utility, and ε_ni is the unobservable portion of utility, which means the effect of unobservable factors on personal utility. For all Q alternative restoration projects, the probability P_ni of respondent n choosing project i rather than project s can be expressed as follows:

$P_{ni}=P(U_{ni}>U_{ns},\forall i\ne s)=P(V_{ni}+{\epsilon }_{ni}>V_{ns}+{\epsilon }_{ns},\forall i\ne s)(2)$

Then, the utility U_ni of a specific restoration project i selected by respondent n can be described as follows:

$U_{ni}={\alpha }_{n}AS{C}_{ni}+{\beta }_n{X}_{ni}+{\gamma }_{n}WT{P}_{ni}+{\epsilon }_{ni}(3)$

Where ASC is the alternative specific constant, which reflects the baseline utility that maintaining the current ecological environment (taking no restoration measurements) can bring to the respondent; X_ni and WTP_ni represent the restoration atrributes of project i and the costs of project i, respectively; ε_ni is a random disturbance term, which is generally assumed to obey the extreme value distribution of type I (i.e., a Gumbel distribution); and α_n, β_n, γ_n reflect the preference degree of respondent n for each restoration attribute, φn = (α_n, β_n, γ_n). Then, the probability distribution function of respondent n choosing project i from all Q situations is (Heiss, 2016):

$P_{ni}=\frac{e^{{\alpha }_{n}AS{C}_{ni}+{\beta }_n{X}_{ni}+{\gamma }_{n}WT{P}_{ni}}}{{\displaystyle \sum _Q{e}^{{\alpha }_{n}AS{C}_{ns}+{\beta }_n{X}_{ns}+{\gamma }_{n}WT{P}_{ns}}}}={\displaystyle \int \frac{e^{V_{ni}\left({\phi }_n\right)}}{{\displaystyle \sum _Q{e}^{V_{nj}\left({\phi }_n\right)}}}f\left({\phi }_n\right)}d{\phi }_n(4)$

In this paper, the RPL model (random parameters logit model) was used for coefficient estimation. This model can estimate each coefficient’s distribution parameters if needed and is not limited to estimating only the coefficients’ fixed mean value (Xu et al., 2020). Therefore, if a specific coefficient is set to conform to a particular distribution (i.e., random parameter), its mean value and standard deviation can be estimated by the RPL model. Otherwise, if it is set as a fixed parameter, only its mean value can be estimated. Therefore, the RPL model has the potential to reveal the variability of respondents’ preferences for different ecological restoration attributes, which is closer to the real-world situation.

Finally, we used the maximum likelihood estimation to obtain each coefficient’s parameters in the above model. Then, the compensation surpluses (CS) of the respondents when they choose different restoration projects (from 1 to 1) can be calculated according to the following formula:

$\begin{array}{l}C{S}_1=(V_0-V_1)/\gamma =[(\alpha +\beta X_0)-\beta X_1]/\gamma \\ C{S}_2=(V_0-V_2)/\gamma =[(\alpha +\beta X_0)-\beta X_2]/\gamma \\ ⋮⋮⋮\\ C{S}_i=(V_0-V_i)/\gamma =[(\alpha +\beta X_0)-\beta X_i]/\gamma \end{array}(5)$

Where α, β, and γ are the average values of each coefficient; V₀ is the status quo without any restoration measurements; and V_i is the value after implementing project i with specific restoration measurements. Therefore, CS_i reflects the respondents’ average WTP for restoration project i and the average social benefits that respondents can obtain from restoration project i.

Study Area

Hainan Province is in the southernmost part of China, and its main landmass is Hainan Island, which is located between 18°10'∼20°10′ north latitude and 108°37'∼111°03′ east longitude and covering an area of approximately 33,900 km². It is the second-largest island in China after the island of Taiwan. Hainan Island’s climate is characterized by a typical oceanic tropical monsoon climate, with high temperatures throughout the year. The average annual temperature is between 22 and 26°C, and the yearly rainfall is abundant with yearly average precipitation above 1,600 mm. With sufficient light and heat resources, many tropical crops are suitable for growing in Hainan, such as rubber, coffee, cocoa, coconut, and palm.

Natural rubber was introduced to Hainan Island in the early 20th century. Subsequently, the planting area gradually expanded and has grown to 520,000 hm², accounting for more than 15% of the total area of Hainan Island. As a result, the original tropical rainforest ecosystem of Hainan Island has been severely damaged, and a series of problems, such as vegetation degradation, soil erosion, a sharp decline in biodiversity, and local droughts have emerged (Xu et al., 2020), which has aroused widespread concern. In April 2018, the central government supports Hainan island in building a pilot free trade zone and supports Hainan in gradually exploring and steadily advancing the construction of a free trade port (FTP), meanwhile “The Guiding Opinions on Supporting Hainan’s Comprehensively Deepening Reform and Opening Up”¹ issued by the CPC Central Committee and the State Council of China stated that the “National Ecological Civilization Pilot Zone (Hainan)” would become one of the main fields for comprehensively deepening the reform and opening up, supporting Hainan by establishing “The Demonstration Province of Circular Ecological Agriculture” and “The Pilot Zone for Green Agricultural Development”. In July 2019, “The Pilot Plan for Hainan Rainforest National Park System”² issued by the National Park Administration stated that more efforts should be made to promote the establishment of Hainan Rainforest National Park, step up ecological protection and restoration, enhance biodiversity conservation, improve water conservation capacity, and promote the formation of a new pattern of harmonious coexistence between humans and nature. Against the strategic background of national ecological civilization construction, the natural rubber planting in Hainan Province urgently needs to change from a traditional operation pattern to an environmentally friendly operation.

Evaluation Attributes

This paper uses ecological attributes and WTP to describe the ecological products (rubber plantation ecological restoration projects) and their cost. Thus, selecting attributes is a prerequisite to implementing a CE. As we aimed to provide useful information for policy design by understanding respondents’ preferences for ecological restoration measures, we focus on the following issues in selecting attributes: how many areas of rubber plantation should be involved? Which regions should be prioritized? Who should be the priority implementers, and what should be done in ecological restoration? In view of this, we chose the attributes through a literature review, a pilot survey, and in-depth discussions with experts in related fields and local government departments to make it representative. Ultimately, four ecological restoration attributes were involved in the final design: restoration area, priority area, priority implementers, and under-forest space utilization. WTP was also included to reveal the respondents’ preferences for the attributes of different ecological restoration projects. In addition, the above attribute levels were set based on historical data, expert consultation, and pilot survey results. Table 1 shows the meaning and level values of the attributes. The selection basis of each attribute and its relation with respondents’ welfare are elaborated as follows:

1) Restoration area. According to the “Work Plan for Delimitation of Rice Production Functional Zones and Natural Rubber Production Protection Zones in Hainan Province” issued by the Hainan Provincial Government in 2017, the delimitation of 8.4 million mu of natural rubber production protection zones is set to complete within 3–5 years³, which will be the total effective area of natural rubber plantation in Hainan province in the future. We assume that ecological restoration projects can cover all rubber plantations if conditions permit. Therefore, we set the maximum restoration area of rubber plantations at 8.4 million mu with 100% coverage. To reveal respondents’ preference for the coverage area of the ecological restoration project, we set another level values of 2.1 million mu, 4.2 million mu and 6.3 million mu according to the isometric principle, which is easy for respondents to understand, with coverage rates of 25, 50, and 75% respectively. It is clear that the ecological restoration of millions of natural rubber plantations is a long process, and it is difficult to achieve the goal in a short time. Therefore, we set another two attributes, namely priority areas and priority implementers, to reveal respondents’ preferences.

2) Priority areas. Existing studies show that excessive rubber plantations can cause severe environmental problems, such as, soil erosions at over-slope locations (Liu et al., 2017, 2015), water pollution and poor water conservation capacity in wellhead protection areas (Guardiola-Claramonte et al., 2008; Tan et al., 2011), biodiversity decrease in biodiversity preservation areas (including nature reserves, national forest parks, etc.) (Li et al., 2006; Ahrends et al., 2015; Warren-Thomas et al., 2015). In addition, rubber forests in or around rural living settlements may also cause some livable rural environment problems, such as reducing the diversity and aesthetics of villages and the ventilation and lighting (Zhang et al., 2019). Respondents may have various preferences for the priority restoration areas mentioned above, so we set four alternatives: over-slope planting area, water source protection area, biodiversity preservation area, and rural living area.

3) Priority implementers. Small householders and state farms are the two main types of rubber growers, which respectively operate 54.85 and 45.15% of rubber plantations in Hainan province by the end of 2017 (Liu et al., 2019). There are many differences in production manners, organizational forms, acquisition of production materials, technical level, and production efficiency between these two types of growers (Zhong et al., 2018). Hence, the promotion effects of ecological restoration projects are likely to differ among different implementers. As beneficiaries, respondents expect the ecological restoration project to be better implemented, we set small householders and state farms as two alternatives for priority implementers.

4) Under-forest space utilization. The development of the under-forest economy is of great significance to reducing biodiversity decrease, soil erosion, ecosystem services degradation, and other impacts caused by rubber planting (Huang et al., 2016; Wen et al., 2018). At present, there are only three relatively mature understory space utilization modes, namely, developing under-forest planting industry, developing under-forest breeding industry, and combining planting and breeding. Respondents may have distinct preferences for under-forest space utilization in the restoration process since each mode can affect species diversity differently. Therefore, we set only Planting, only Breeding, and Planting & Breeding as alternatives.

5) WTP. The fifth attribute, willingness to pay, is a household payment to improve the rubber plantation ecosystem. In addition to the cash form of WTP, it can also be presented as an increase in the cost of living due to a rise in environmental protection taxes and fees or indirect costs caused by ecological protection and reduced economic development (Alcon et al., 2020; Hynes et al., 2021). The payment levels were determined through an open-ended CVM in a pre-survey. The payment levels are 25, 50, 75, and 100 yuan, and if the respondents choose “no restoration measures”, they do not need to pay.

TABLE 1

TABLE 1. Attributes of rubber plantation ecological restoration.

Questionnaire Design

The selected attributes and levels were combined into several choice sets, and this produces so many alternatives that it would be overly cumbersome and intellectually demanding for respondents to choose among them. Researchers generally design about three choice sets in each questionnaire and three alternatives in each choice set. For the sake of experimental effect, each of our CE questionnaires provided the respondents with two choice sets, and each choice set contained three alternatives: “Project A”, “Project B”, and “Project C” (shown in Figure 2). “Project A” and “Project B″ comprise different restoration measures that are represented by the attribute levels, and “Project C” is an alternative without any restoration measures, and the respondents do not need to pay for it⁴. Hence based on the attributes and their level values shown in Table 1, there would be 384 possible alternatives (4*4*2*3*4)⁵, 73,536 possible choice sets ($C_{384}^2$), and 6.627*10¹³possible CE questionnaires ($C_{73536}^3$). If all the possible CE questionnaires were tested in the field survey, it would take much work and material resources. After determining the number of alternatives, choice sets, and questionnaires, we used Ngene1.1.1 to conduct an orthogonal experimental design, and representative CE questionnaires were selected based on orthogonality. Finally, eight versions of CE questionnaires containing sixteen choice sets were generated. The efficiency measure results of the orthogonal experiment had a D-error of 0.002194 and an A-error of 0.050781⁶. Then, we checked the rationality of each choice set and adjusted the choice sets with the dominant strategy.

FIGURE 2

FIGURE 2. Example of choice sets in the CE questionnaires.

A prerequisite for successfully implementing CE is respondents can understand the questionnaire accurately. Thus following measures were taken to ensure the validity of the experiments. First, we further optimized the expression of each attribute to make sure every respondent could understand the meaning of each attribute. Respondents were required to be presented with a detailed description of the four attributes and informed with survey questions. Second, our CE questionnaires used text and graphics (Figure 2), which made the restoration projects more vivid and added to the fun of the experiments to make them easier for respondents to understand. Third, each questionnaire contained only two choice sets to avoid fatigue due to repeated experiments. Fourth, a dual error control mechanism was introduced to judge the experiments’ validity, and the respondents were required to evaluate their understanding and attitude toward the experiments, and the investigators needed to assess the respondent’s cooperation and seriousness. Fifth, all investigators were professionally trained and informed of the experimental operation process and seriousness in detail to ensure the validity of the experiments before the final survey.

Data Collection

Research data were obtained from a field survey of urban residents conducted in Hainan Province in July 2019. Questionnaires were distributed through random sampling, stratified by population size, socioeconomic development level, and rubber plantation location. Four cities where rubber plantations are surrounded were selected, namely Haikou, Sanya, Danzhou, and Wanning, located around Hainan Island. Then 2-3 districts were randomly selected from each city, and respondents were randomly selected from the main streets of each district. A questionnaire survey was conducted through face-to-face interviews. Before filling out the questionnaire formally, the interviewer explained the purpose of the survey and the operation process of the CE in detail to the interviewees to ensure the authenticity and validity of responses. Finally, a total of 550 questionnaires were given out, and 518 were valid after inspection of invalid ones, accounting for an effective response rate of 94.18%. Table 2 shows the socio-demographic characteristics of the respondents.

TABLE 2

TABLE 2. Descriptive statistics of the sample.

Results

Variable Descriptive Statistics

Table 3 provides an overview of the mean summary statistics for attribute values of the rubber plantation restoration projects which the respondents selected. The average restoration area was 1.9672, indicating that the selected projects’ average restoration area was close to 4.2 million mu. Regarding the priority areas, the over-slope planting area’s value was 0.2259, which was the highest, indicating that the respondents were more supportive of promoting ecological restoration projects in this area. For priority implementers, small householders and state farms accounted for 38.22 and 41.12%, respectively. More respondents choose to develop planting and breeding (28.28%) under-forest, slightly higher than only breeding (26.06%) and only planting (25.00%). Additionally, the average ASC value was 0.2066, indicating that the probability of respondents choosing “Project C” was approximately 20%. The statistical results of WTP show that respondents were willing to pay for ecological restoration projects, and the average value was 45.22 yuan.

TABLE 3

TABLE 3. Descriptive statistics of variables.

RPL Model Estimation and Public Preferences Analysis

Table 4 presents the results from the RPL model estimated by Stata 15.0. The dependent variable was the respondent’s choice, and the independent variables included random parameter variables and fixed parameter variables. The variables of restoration area, priority areas, priority implementers, and under-forest space utilization were set as random parameters to estimate their location parameters and scale parameter, that is, the mean and standard deviation of their coefficients. ASC and WTP were set as fixed parameters, and we could only estimate their location parameter. The results showed that the chi-square statistic reached a significance level of 1% for the model’s overall fit, indicating that the econometric model was statistically significant overall.

TABLE 4

TABLE 4. RPL model estimation results.

In terms of the fixed parameter variables, it showed that the mean value of the ASC coefficient was significantly negative at the level of 1%, indicating that respondents preferred to reject “Project C”. The mean value of the WTP coefficient was significantly negative, indicating that WTP was negatively related to utility. It means that respondents were willing to acquire more ecological improvements with fewer fees, which is consistent with related studies (Cai, et al.; Yao, et al., 2018). In the random parameter variables, the mean value of the restoration area was statistically significantly at the 1% level, indicating that increasing the ecological restoration area reduced the respondents’ utility. According to the field surveys, we also learned that, although most of the respondents supported rubber plantation ecological restoration projects, they did not prefer projects with a larger restoration area because of concern about the negative impact on rubber production and the income of rubber growers. Nevertheless, respondents hoped to prioritize ecological restoration projects in areas they consider critical. The estimation results based on rural living areas showed that mean values of the over-slope planting area were significant at 5%, and the water source protection area was significant at 1%, respectively. It indicated that prioritizing ecological restoration projects in these two areas is more desirable to the respondents than prioritizing projects in rural living areas. The estimated result of the biodiversity preservation area was not significant, indicating that there was no difference in prioritizing ecological restoration projects in this area and rural living areas in terms of the respondents’ utility. This result may be related to the residents’ low ecological awareness. In the field survey, we found that many respondents still do not realize the ecological problems caused by substituting rubber plantations for natural forests. Some even thought that rubber trees were good for the local ecology because they are green. For priority implementers, the estimated results based on state farms showed that the mean values of the small householders were not statistically significant. It indicates that respondents did not have obvious preferences for prioritizing ecological restoration projects with small householders than state farms. This is possible because those respondents hope both state farms and smallholders can be equally involved in ecological restoration projects. They pay more attention to the implementation scale and location of the project but do not care about who will implement the project. As mentioned above, there are obvious differences between the two types of rubber farmers in many aspects, and the future ecological restoration policy design needs to be differentiated and targeted. In terms of under-forest space utilization, the estimated results based on only planting showed that the mean values of only breeding and planting and breeding were both statistically significant at 10%. It indicates that respondents have a significant preference for only planting and breeding or planting and breeding. The reason may be that some respondents believe that the economic benefits of simply developing an under-forest planting industry may not be high enough to ensure the sustainability of the restoration project.

For the standard deviation of random parameters, the coefficient of restoration area was statistically significant at the 1% level, which means respondents have considerable heterogeneity in preference for the scale of the restored area. The coefficient of the over-slope planting area and the biodiversity preservation area were significant at the 1% level. In contrast, the water source protection area’s coefficient was not significant. It indicated considerable heterogeneity existed in the respondents’ preferences for the regions of over-slope and biodiversity preservation (relative to rural living areas). There was no significant heterogeneity in the preferences for water source protection areas. The coefficient of small householders was not significant, indicating no significant heterogeneity in the respondents’ preferences for small householders (relative to state farms). The coefficients of only planting and planting and breeding were significant at the 1% level, indicating that the respondents’ preferences for these two utilization forms (relative to only planting) were significantly heterogeneous. This result also suggested that the assumption of preference heterogeneity with the RPL model was closer to the actual situation (Hensher and Greene, 2002; Birol et al., 2006; Hynes et al., 2021).

Social Benefits Calculation

The compensation surplus (CS) can reflect the average WTP of respondents to improve the rubber plantations’ ecological condition, that is, the WTP for the social benefits that can be achieved when the rubber plantations’ ecological condition is enhanced from the current to a higher level. Generally, the higher the respondents’ WTP, the greater the CS, indicating that improving the ecological condition will bring more social benefits. According to the previous analysis and the principle of maximizing social benefits, we made attribute level selection for potential rubber plantation ecological restoration projects design: First, since the mean value of the restoration area’s coefficient was negative, indicating that the respondents did not want to restore too much area currently; thus we take relatively low restoration area, 2.1 million mu (25% coverage) and 4.2 million mu (50% coverage) involved. Second, respondents were more inclined to the over-slope planting area and water source protection area; therefore, these two types of areas should be included. Third, the respondents had no significant preference differences between small householders and state farms; thus, both these two implementers should be considered in the potential projects. Fourth, the respondents were more inclined to only breeding and planting and breeding in terms of the under-forest space utilization, so these two utilization forms should be taken into account.

A total of sixteen potential restoration projects were created based on the attributes and levels selected above (shown in Table 5), and the CS of each project can be calculated separately. Project 4 has the lowest annual CS of 232.7912 yuan/household, and project 13 has the highest annual CS of 320.8280 yuan/household. We further calculated the total social benefits (TSB) that each ecological restoration project could bring through the sum of individual CS for the whole target population. By querying the “Hainan Statistical Yearbook (2019)⁷”, we confirm that the total number of urban households in Hainan Province was 1.0521 million at the end of 2018. Then, we calculated that project had the lowest TSB, at 244.9196 million yuan/year, and project five had the highest TSB, at 337.5431 million yuan/year. The results can be used as the budget basis for implementing the ecological restoration project of rubber plantations in Hainan Province. It also indicates that the public demands these ecological products provided by the government, and also they can accept the corresponding financial expenditure. These budget funds could be used for many aspects of rubber management in the future, such as natural forest reconstruction in critical regions (such as over-slope planting area, water source protection area), the corresponding farmers’ compensation fees, diversity of planting and breeding technology promotion, promotion or subsidy of green pollution-free technology related to rubber production, ecological propaganda and education for farmers, etc.

TABLE 5

TABLE 5. Compensation surplus and total social benefits of different potential restoration projects.

Discussion

Implementing the rubber plantation ecological restoration project is of great significance to improve the ecological service function of tropical rain forests and the ecosystem of Hainan Island. However, benefits of ecological restoration projects for rubber plantations, such as soil and water conservation, increased biodiversity, and possible recreational opportunities, are not generally traded on the market and therefore do not generally have a clear market price. Failure to incorporate these non-market values into the decision-making process may lead us to make decisions not actually in society’s interest (Boyer and Polasky, 2009; Frynas et al., 2017). In addition, effectively connecting the policy design of ecological restoration with the public preference is also a topic worthy of attention, related to whether the corresponding policy design can obtain enough public support. The contribution of this study is to help policymakers get the social benefits of non-market rubber plantation restoration projects and understand the public preference for ecological restoration policies. The results indicate a positive and significant societal willingness to pay for the ecological restoration of natural rubber plantation in Hainan Province, which is consistent with prior studies that have valued other types of ecosystem restoration activities (Lan et al., 2017; Jin, et al., 2018; Xu, et al., 2020; Shi et al., 2021). In addition, there are several aspects worthy of further discussion for the current work.

1) The benefit we measured only reflects the support degree of urban residents in Hainan province for the rubber plantation ecological restoration project and does not include rubber farmers. In fact, rubber farmers will also enjoy ecological benefits from implementing the ecological restoration project. There are two reasons why rubber farmers are not included. On the one hand, as the economic basis of farmers is relatively weak, they rarely can pay for ecological products. Additionally, most farmers are poorly educated, and their ecological cognition are comparatively insufficient, so it is pretty difficult for them to understand the experiments. On the other hand, rubber farmers are both beneficiaries and victims of ecological restoration projects, making it complicated for them to weigh their interests and provide us with accurate information about their willingness to pay. It is also necessary for us to adopt a more appropriate cost-benefit quantitative analysis method in further research to explore the impact of the ecological restoration project on rubber farmers to improve the accuracy of benefit calculation.

2) The estimated coefficient of ecological restoration area is negative, indicating that residents do not want to implement the ecological restoration project on a large scale. As mentioned above, the respondents may be concerned that the implementation of ecological projects will affect the economic income of rubber farmers. As we know, rubber planting is one of the important economic sources for local farmers. In the past 10 years, natural rubber prices have continued to maintain low, significantly impacting rubber growers’ income. It also implies that the future ecological restoration policy design needs to pay attention to the sustainable livelihood of rubber farmers. Supporting policies and measures such as subsidies (ecological compensation), skills training, entrepreneurship guidance and credit support can be provided to make up for farmers’ economic losses caused by the ecological restoration projects. Moreover, future research may consider restoration projects’ influence on rubber farmers’ production and life, and corresponding attributes can be involved. We can include the aforementioned supporting policies and measures as one of the attributes in the CE questionnaire to minimize respondents’ concern on this issue.

3) We also found that the respondents did not have a sufficient and unified understanding of the ecological and environmental problems caused by rubber planting. Some respondents believe that rubber trees are green and don’t cause any ecological problems. This is the actual logic of many respondents, which may affect the experiment results or even hinder the experiment by significantly increasing the proportion of respondents who protest payment. To solve this problem, we set up a warm-up session before the choice experiments to judge the respondents’ awareness through simple questions about ecological problems caused by rubber planting. Then, before the formal experiments, strictly trained investigators were required to explain to the respondents the ecological and environmental problems brought by the development of rubber plantations. In this way, we try to make each respondent make choices with consistent background information. It also inspired us to the importance of doing preliminary research, finding potential problems, and taking effective solutions.

4) The research still has two deficiencies in experimental design and model analysis. Firstly, efficient design is a commonly used method in orthogonal experimental design, requiring some prior information. Still, other studies have shown that the random design (which is the easiest to generate) performs as well as the efficient design, and even better if data cleaning is done to remove choice tasks where one alternative dominates the other (Walker et al., 2017). Due to the lack of related studies on the ecological governance of rubber plantations and the tight schedule, we did not obtain enough prior information to conduct an efficient design. In subsequent studies, if prior information cannot be obtained from existing studies, we should try to get prior information through a small range of pre-research and optimize the experimental design based on pre-research data. Secondly, we adopted the RPL model, which assumed that respondents’ preferences were heterogeneous and closer to reality, making estimation results for the evaluation attribute coefficients more accurate and reducing model setting errors. However, we focus on analyzing preference differences at the group level and do not further discuss the heterogeneity of preferences among individuals. Subsequent studies can use the Latent Class model (LCM) to analyze further the impact of individual preference differences on benefit evaluation (Strazzera et al., 2012; Barrio and Loureiro, 2018).

Conclusions

Against the background of constructing an FTP on the whole island of Hainan, the National Ecological Civilization Pilot Zone (Hainan) holds a new strategic position. The state and the local government have successively issued policies and measures to break resource constraints, reduce environmental pollution, and ameliorate ecosystem degradation in Hainan Province. In response to the ecological problems caused by natural rubber planting, accelerating rubber plantation ecological restoration has become the key to improving the ecological quality and sustainability of the natural rubber planting industry.

Based on a CE field survey, this research presented rubber plantation ecological restoration projects to respondents as ecological products and simulated urban residents’ ecological consumption behavior in Hainan Province. Then, we analyzed respondents’ preferences with different restoration attributes using the RPL model and calculated the social benefits brought by the potential restoration scenarios. The main findings of this study were as follows: first, urban residents in Hainan Province showed a positive attitude toward rubber plantation ecological restoration. They were willing to bear a certain amount of cost to improve the ecological condition of rubber plantations. Second, the respondents had a significant preference for different restoration attributes. For example, in terms of the restoration area, they were not inclined to feel overly involved. Respondents preferred priority restoration in over-slope planting areas and water source protection areas. For under-forest space utilization, they were more inclined to only breeding and planting and breeding. Third, by offering a variety of potential restoration projects, we found that a project covering 2.1 million mu of rubber plantations, prioritizing the water source protection area, focusing the implementation on state farms and developing the under-forest breeding economy would gain more social benefits, precisely, 337.543 million yuan/year.

In response to the findings, we propose the following policy recommendations. First, costs and benefits, especially non-market social benefits, should be considered when designing or implementing rubber plantation ecological restoration projects, which can be calculated by residents’ welfare improved by implementing rubber ecological plantation projects. Still, it is challenging to quantify these welfare improvements. At present, some non-market valuation methods, such as CVM and CE, can provide feasible approaches, but undeniably, these methods are still being improved. Second, ecological restoration policies should consider public needs more and incorporate them into future guidelines, which will undoubtedly enhance the accuracy of public environmental policies and the effectiveness of their implementation, as long as we can reveal public preferences well. Third, rubber plantation ecological restoration should focus on restoration quality rather than blindly aiming at the restoration area to maximize social benefits. In the process, priority should be given to selecting rubber plantations close to the water source protection area, small householders as the priority implementers, and an appropriate under-forest economy.

Data Availability Statement

The data analyzed in this study is subject to the following licenses/restrictions: The data are not publicly available due to personal privacy and non-open access of the research program. Requests to access these datasets should be directed to Xu Tao, xutao_2013@outlook.com.

Ethics Statement

Ethical review and approval was not required for the study on human participants in accordance with the local legislation and institutional requirements. The patients/participants provided their written informed consent to participate in this study.

Author Contributions

DQ and TX initiated the study concept. DQ, WL, TX led the data collection. TX conducted data analysis and interpretation of the data. DQ wrote the first draft of the manuscript. DZ and YY contributed to language control and revised the manuscript.

Funding

This study was supported by the National Natural Science Foundation of China (No.72003054, No. 72103052), Humanities and Social Sciences Program of the Ministry of Education (No. 19XJC790011), Hainan Provincial Natural Science Foundation of China (No. 719QN197), National Natural Rubber Industry Technical System Industrial Economic Position (No. CARS-33-CJ1), and Natural Rubber Industry Operation Early-warning System Construction Project (No. 18210024). Hainan Provincial Natural Science Foundation of China (No. 719QN198).

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s Note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

¹Source: http://www.gov.cn/zhengce/2018-04/14/content_5282456.htm

²Source: https://www.hainan.gov.cn/hainan/zchbbwwj/202008/f0a42020ac1547098d502acd161119cf.shtml

³Source: http://www.gov.cn/xinwen/2017-11/23/content_5241750.htm

⁴According to the research (Adamowicz et al., 1998), if the “do not participate” option is omitted, it is difficult for respondents to make effective choices when all restoration projects are unattractive. Therefore, “Project C” with “No restoration, no payment” was added to the questionnaire in this article to indicate that the respondents do not support any of the above alternative restoration projects

⁵The level of each attribute corresponding to “0” WTP is a fixed value, that is, no restoration measures are taken. Therefore, only four levels of the WTP were entered into the calculation

⁶Both the D-error and the A-error were derived from the progressive variance covariance matrix. The D-error takes the determinant of the matrix, and the A-error takes the trace of the matrix. The ideal experimental design can achieve the minimization of the two

⁷Source: http://stats.hainan.gov.cn/tjj/tjsu/

References

Adamowicz, W., Boxall, P., Williams, M., and Louviere, J. (1998). Stated Preference Approaches for Measuring Passive Use Values: Choice Experiments and Contingent Valuation. Am. J. Agric. Econ. 80, 64–75. doi:10.2307/3180269

CrossRef Full Text | Google Scholar

Ahlheim, M., Börger, T., and Frör, O. (2015). Replacing Rubber Plantations by Rain forest in Southwest China-who Would Gain and How Much. Environ. Monit. Assess. 187, 3. doi:10.1007/s10661-014-4088-8

PubMed Abstract | CrossRef Full Text | Google Scholar

Ahrends, A., Hollingsworth, P. M., Ziegler, A. D., Fox, J. M., Chen, H., Su, Y., et al. (2015). Current Trends of Rubber Plantation Expansion May Threaten Biodiversity and Livelihoods. Glob. Environ. Change 34, 48–58. doi:10.1016/J.GLOENVCHA.2015.06.002

CrossRef Full Text | Google Scholar

Alcon, F., Marín-Miñano, C., Zabala, J. A., de-Miguel, M.-D., and Martínez-Paz, J. M. (2020). Valuing Diversification Benefits through Intercropping in Mediterranean Agroecosystems: A Choice experiment Approach. Ecol. Econ. 171, 106593. doi:10.1016/j.ecolecon.2020.106593

CrossRef Full Text | Google Scholar

Barrio, M., and Loureiro, M. L. (2018). Evaluating Management Options for a Marine and Terrestrial National Park: Heterogeneous Preferences in Choice Experiments. Mar. Pol. 95, 85–94. doi:10.1016/J.MARPOL.2018.06.015

CrossRef Full Text | Google Scholar

Bingli, L., Weide, Z., and Rongyuan, Z. (2020). The Rebirth of Tropical Rainforest — Ecological Restoration Planning for Sanda Mountain of Xishuangbanna, China. Landsc. Archit. Front. 8, 108. doi:10.15302/j-laf-1-040012

CrossRef Full Text | Google Scholar

Birol, E., Karousakis, K., and Koundouri, P. (2006). Using a Choice experiment to Account for Preference Heterogeneity in Wetland Attributes: The Case of Cheimaditida Wetland in Greece. Ecol. Econ. 60, 145–156. doi:10.1016/j.ecolecon.2006.06.002

CrossRef Full Text | Google Scholar

Blanco, E., Haller, T., and Walker, J. M. (2018). Provision of Environmental Public Goods: Unconditional and Conditional Donations from Outsiders. J. Environ. Econ. Manage. 92, 815–831. doi:10.1016/j.jeem.2017.10.002

CrossRef Full Text | Google Scholar

Blanco, E., Struwe, N., and Walker, J. M. (2021). Experimental Evidence on Sharing Rules and Additionality in Transfer Payments. J. Econ. Behav. Organ. 188, 1221–1247. doi:10.1016/j.jebo.2021.06.012

CrossRef Full Text | Google Scholar

Boyer, T., and Polasky, S. (2009). Valuing Urban Wetlands: A Review of Non-market Valuation Studies. Wetlands 24 (4), 744–755. doi:10.1672/0277-5212(2004)024[0744:vuwaro]2.0.co;2

CrossRef Full Text | Google Scholar

Cai, Y., Zhao, M., Shi, Y., and Khan, I. (2020). Assessing Restoration Benefit of Grassland Ecosystem Incorporating Preference Heterogeneity Empirical Data from Inner Mongolia Autonomous Region. Ecol. Indicators 117, 106705. doi:10.1016/j.ecolind.2020.106705

PubMed Abstract | CrossRef Full Text | Google Scholar

Chen, H., Yi, Z.-F., Schmidt-Vogt, D., Ahrends, A., Beckschäfer, P., Kleinn, C., et al. (2016). Pushing the Limits: The Pattern and Dynamics of Rubber Monoculture Expansion in Xishuangbanna, SW China. PloS one 11, e0150062. doi:10.1371/journal.pone.0150062

PubMed Abstract | CrossRef Full Text | Google Scholar

Delaney, J., and Jacobson, S. (2014). Those Outsiders: How Downstream Externalities Affect Public Good Provision. J. Environ. Econ. Manage. 67, 340–352. doi:10.1016/J.JEEM.2013.12.007

CrossRef Full Text | Google Scholar

Frynas, J. G., Child, J., and Tarba, S. Y. (2017). Non-market Social and Political Strategies - New Integrative Approaches and Interdisciplinary Borrowings. Br. J Manage 28, 559–574. doi:10.1111/1467-8551.12253

CrossRef Full Text | Google Scholar

Gan, Y., Xu, T., Xu, N., Xu, J., and Qiao, D. (2021). How Environmental Awareness and Knowledge Affect Urban Residents' Willingness to Participate in Rubber Plantation Ecological Restoration Programs: Evidence from Hainan, China. Sustainability 13, 1852. doi:10.3390/SU13041852

CrossRef Full Text | Google Scholar

Gashu, K., and Aminu, O. (2019). Participatory forest Management and Smallholder Farmers' Livelihoods Improvement Nexus in Northwest Ethiopia. J. Sustain. For. 38, 413–426. doi:10.1080/10549811.2019.1569535

CrossRef Full Text | Google Scholar

Grist, P., and Menz, K. (1995). Modelling the Economics of Imperata Control in Smallholder Rubber Plantations. undefined.

Google Scholar

Guardiola-Claramonte, M., Troch, P. A., Ziegler, A. D., Giambelluca, T. W., Vogler, J. B., and Nullet, M. A. (2008). Local Hydrologic Effects of Introducing Non-native Vegetation in a Tropical Catchment. Ecohydrol. 1, 13–22. doi:10.1002/ECO.3

CrossRef Full Text | Google Scholar

Hammond, J., van Wijk, M. T., Smajgl, A., Ward, J., Pagella, T., Xu, J., et al. (2017). Farm Types and Farmer Motivations to Adapt: Implications for Design of Sustainable Agricultural Interventions in the Rubber Plantations of South West China. Agric. Syst. 154, 1–12. doi:10.1016/J.AGSY.2017.02.009

CrossRef Full Text | Google Scholar

Heiss, F. (2016). Discrete Choice Methods with Simulation. Econometric Rev. 35, 688–692. doi:10.1080/07474938.2014.975634

CrossRef Full Text | Google Scholar

Hensher, D., and Greene, W., 2002. The Mixed Logit Model: The State of Practice. doi:10.1023/A:1022558715350

CrossRef Full Text | Google Scholar

Himes-Cornell, A., Pendleton, L., and Atiyah, P. (2018). Valuing Ecosystem Services from Blue Forests: A Systematic Review of the Valuation of Salt Marshes, Sea Grass Beds and Mangrove Forests. Ecosystem Serv. 30, 36–48. doi:10.1016/J.ECOSER.2018.01.006

CrossRef Full Text | Google Scholar

Huang, X., Lan, G., Yang, C., Wu, Z., and Tao, Z. (2016). Shrub-grass Species Diversity of Rubber Plantations under Different Cultivation Patterns in Hainan. J. Northwest For. Univ. 31, 115–120.

Google Scholar

Huang, X., and Kim, S. E. (2020). When Top‐down Meets Bottom‐up: Local Adoption of Social Policy Reform in China. Governance 33, 343–364. doi:10.1111/gove.12433

CrossRef Full Text | Google Scholar

Hynes, S., Chen, W., Vondolia, K., Armstrong, C., and O'Connor, E. (2021). Valuing the Ecosystem Service Benefits from Kelp forest Restoration: A Choice experiment from Norway. Ecol. Econ. 179, 106833. doi:10.1016/j.ecolecon.2020.106833

CrossRef Full Text | Google Scholar

Jin, J., He, R., Wang, W., and Gong, H. (2018). Valuing Cultivated Land protection: A Contingent Valuation and Choice experiment Study in China. Land Use Policy 74, 214–219. doi:10.1016/J.LANDUSEPOL.2017.09.023

CrossRef Full Text | Google Scholar

Johnston, R. J. (2007). Choice Experiments, Site Similarity and Benefits Transfer. Environ. Resource Econ. 38, 331–351. doi:10.1007/S10640-006-9073-4

CrossRef Full Text | Google Scholar

Knickel, K., and Maréchal, A. (2018). Stimulating the Social and Environmental Benefits of Agriculture and Forestry: An EU-Based Comparative Analysis. Land Use Policy 73, 320–330. doi:10.1016/J.LANDUSEPOL.2017.12.064

CrossRef Full Text | Google Scholar

Lancaster, K. (1966). A New Approach to Consumer Theory. J. Econ. Theor. 32 (1), 93–110. doi:10.1016/0022-0531(84)90076-0

CrossRef Full Text | Google Scholar

Lan, J., Xia, W., Liu, L., and Ou, W. (2017). Public Preferences for Biological Resource Conservation Based on Choice Experiment Methods. Resources Science 39 (3), 577–584. doi:10.18402/resci.2017.03.19

CrossRef Full Text | Google Scholar

Li, H., Aide, T. M., Ma, Y., Liu, W., and Cao, M. (2006). “Demand for Rubber Is Causing the Loss of High Diversity Rain forest in SW China,” in Plant Conservation and Biodiversity, Topics in Biodiversity and Conservation. Editors D.L. Hawksworth, and A.T. Bull (Dordrecht: Springer Netherlands), 157–171. doi:10.1007/978-1-4020-6444-9_11

CrossRef Full Text | Google Scholar

Liu, R., Mo, Y., Yang, L., He, C., Fu, Li., Wu, W., et al. (2019). Scientific and Technological Innovation Chain Promotes the Development of Hainan Natural Rubber Industry Chain. Sci. Tech. Manage. Res. 39, 91–97.

Google Scholar

Liu, W., Luo, Q., Li, J. T., Wang, P., Lu, H., Liu, W.-Y., et al. (2015). The Effects of Conversion of Tropical Rainforest to Rubber Plantation on Splash Erosion in Xishuangbanna. SW China: undefined.

Google Scholar

Liu, W., Luo, Q., Lu, H., Wu, J., and Duan, W. (2017). The Effect of Litter Layer on Controlling Surface Runoff and Erosion in Rubber Plantations on Tropical Mountain Slopes, SW China. Catena 149, 167–175. doi:10.1016/J.CATENA.2016.09.013

CrossRef Full Text | Google Scholar

Liu, X., Feng, Z., Jiang, L., Li, P., Liao, C., Yang, Y., et al. (2013). Rubber Plantation and its Relationship with Topographical Factors in the Border Region of China, Laos and Myanmar. J. Geogr. Sci. 23, 1019–1040. doi:10.1007/s11442-013-1060-4

CrossRef Full Text | Google Scholar

Ma, X., Lacombe, G., Harrison, R., Xu, J., and van Noordwijk, M. (2019). Expanding Rubber Plantations in Southern China: Evidence for Hydrological Impacts. Water 11, 651. doi:10.3390/w11040651

CrossRef Full Text | Google Scholar

McFadden, D. (1973). “Conditional Logit Analysis of Qualitative Choice Behavior,” in Frontiers in Econometrics. Editor P. Zarembka (New York, NY, USA: Academic Press).

Google Scholar

Min, S., Bai, J., Huang, J., and Waibel, H. (2018). Willingness of Smallholder Rubber Farmers to Participate in Ecosystem protection: Effects of Household Wealth and Environmental Awareness. For. Pol. Econ. 87, 70–84. doi:10.1016/J.FORPOL.2017.11.009

CrossRef Full Text | Google Scholar

Min, S., Huang, J., Waibel, H., Yang, X., and Cadisch, G. (2019). Rubber Boom, Land Use Change and the Implications for Carbon Balances in Xishuangbanna, Southwest China. Ecol. Econ. 156, 57–67. doi:10.1016/J.ECOLECON.2018.09.009

CrossRef Full Text | Google Scholar

Nur Syuhada, C. I., Mahirah, K., and Roseliza, M. A. (2020). Dealing with Attributes in a Discrete Choice experiment on Valuation of Water Services in East Peninsular Malaysia. Utilities Policy 64, 101037. doi:10.1016/j.jup.2020.101037

CrossRef Full Text | Google Scholar

Olsen, S. B., Jensen, C. U., and Panduro, T. E. (2019). Modelling Strategies for Discontinuous Distance Decay in Willingness to Pay for Ecosystem Services. Environ. Resource Econ. 75, 351–386. doi:10.1007/s10640-019-00370-7

CrossRef Full Text | Google Scholar

Ouyang, J., Zhang, K., Wen, B., and Lu, Y. (2020). Top-Down and Bottom-Up Approaches to Environmental Governance in China: Evidence from the River Chief System (RCS). Ijerph 17, 7058. doi:10.3390/ijerph17197058

PubMed Abstract | CrossRef Full Text | Google Scholar

Petcharat, A., Lee, Y., and Chang, J. B. (2020). Choice Experiments for Estimating the Non-market Value of Ecosystem Services in the Bang Kachao Green Area, Thailand. Sustainability 12, 7637. doi:10.3390/su12187637

CrossRef Full Text | Google Scholar

Ruto, E., and Garrod, G. (2009). Investigating Farmers' Preferences for the Design of Agri-Environment Schemes: a Choice experiment Approach. J. Environ. Plann. Manage. 52, 631–647. doi:10.1080/09640560902958172

CrossRef Full Text | Google Scholar

Schaafsma, M., Brouwer, R., and Rose, J. (2012). Directional Heterogeneity in WTP Models for Environmental Valuation. Ecol. Econ. 79, 21–31. doi:10.1016/J.ECOLECON.2012.04.013

CrossRef Full Text | Google Scholar

Smajgl, A., Xu, J., Egan, S., Yi, Z.-F., Ward, J., and Su, Y. (2015). Assessing the Effectiveness of Payments for Ecosystem Services for Diversifying Rubber in Yunnan, China. Environ. Model. Softw. 69, 187–195. doi:10.1016/j.envsoft.2015.03.014

CrossRef Full Text | Google Scholar

Shi, Y., Li, C., and Zhao, M. (2021). Herders’ Aversion to Wildlife Population Increases in Grassland Ecosystem Conservation: Evidence From a Choice Experiment Study. Glob. Ecol. Conserv. 30, e01777. doi:10.1016/j.gecco.2021.e01777

CrossRef Full Text | Google Scholar

Strazzera, E., Mura, M., and Contu, D. (2012). Combining Choice Experiments with Psychometric Scales to Assess the Social Acceptability of Wind Energy Projects: A Latent Class Approach. Energy Policy 48, 334–347. doi:10.1016/J.ENPOL.2012.05.037

CrossRef Full Text | Google Scholar

Sturgeon, J., Menzies, N., and Schillo, N. (2014). Ecological Governance of Rubber in Xishuangbanna, China. Conservat Soc. 12, 376. doi:10.4103/0972-4923.155581

CrossRef Full Text | Google Scholar

Tan, Z.-H., Zhang, Y.-P., Song, Q.-H., Liu, W.-J., Deng, X.-B., Tang, J.-W., et al. (2011). Rubber Plantations Act as Water Pumps in Tropical China. Geophys. Res. Lett. 38, a–n. doi:10.1029/2011GL050006

CrossRef Full Text | Google Scholar

Ureta, J., Motallebi, M., Vassalos, M., Alhassan, M., and Ureta, J. C. (2021). Valuing Stakeholder Preferences for Environmental Benefits of Stormwater Ponds: Evidence from Choice experiment. J. Environ. Manage. 293, 112828. doi:10.1016/j.jenvman.2021.112828

CrossRef Full Text | Google Scholar

Walker, J. L., Wang, Y., Thorhauge, M., and Ben-Akiva, M. (2017). D-Efficient or Deficient? A Robustness Analysis of Stated Choice Experimental Designs. Theor. Decis 84, 215–238. doi:10.1007/S11238-017-9647-3

CrossRef Full Text | Google Scholar

Warren-Thomas, E., Dolman, P. M., and Edwards, D. P. (2015). Increasing Demand for Natural Rubber Necessitates a Robust Sustainability Initiative to Mitigate Impacts on Tropical Biodiversity. Conservation Lett. 8, 230–241. doi:10.1111/CONL.12170

CrossRef Full Text | Google Scholar

Wen, Z., Zhao, H., Liu, L., Li, Y., Mi, H., Ouyang, Z., et al. (2018). Effects of Intercropping with Alpinia Oxyphylla in Rubber Plantation on Soil Water Conserva-Tion Function. Chin. J. Ecol. 37, 3179–3185.

Google Scholar

Widianingsih, N. N., David, W., Pouliot, M., and Theilade, I. (2019). Land Use, Income, and Ethnic Diversity in the Margins of Hutan Harapan - A Rainforest Restoration Concession in Jambi and South Sumatra, Indonesia. Land Use Policy 86, 268–279. doi:10.1016/J.LANDUSEPOL.2019.05.006

CrossRef Full Text | Google Scholar

Wigboldus, S., Hammond, J., Xu, J., Yi, Z.-F., He, J., Klerkx, L., et al. (2017). Scaling green Rubber Cultivation in Southwest China-An Integrative Analysis of Stakeholder Perspectives. Sci. Total Environ. 580, 1475–1482. doi:10.1016/j.scitotenv.2016.12.126

PubMed Abstract | CrossRef Full Text | Google Scholar

Wu, Z.-L., Liu, H.-M., and Liu, L.-Y. (2001). Rubber Cultivation and Sustainable Development in Xishuangbanna, China. Int. J. Sustain. Dev. World Ecol. 8, 337–345. doi:10.1080/13504500109470091

CrossRef Full Text | Google Scholar

Xiao, C., Li, P., and Feng, Z. (2019). Monitoring Annual Dynamics of Mature Rubber Plantations in Xishuangbanna during 1987-2018 Using Landsat Time Series Data: A Multiple Normalization Approach. Int. J. Appl. Earth Observation Geoinformation 77, 30–41. doi:10.1016/j.jag.2018.12.006

CrossRef Full Text | Google Scholar

Xu, J., van Noordwijk, M., He, J., Kim, K.-J., Jo, R.-S., Pak, K.-G., et al. (2012). Participatory Agroforestry Development for Restoring Degraded Sloping Land in DPR Korea. Agroforest Syst. 85, 291–303. doi:10.1007/s10457-012-9501-0

CrossRef Full Text | Google Scholar

Xu, T., Ni, Q., Yao, L.-y., Qiao, D., and Zhao, M.-j. (2020). Public Preference Analysis and Social Benefits Evaluation of River Basin Ecological Restoration: Application of the Choice Experiments for the Shiyang River, China. Discrete Dyn. Nat. Soc. 2020, 1–12. doi:10.1155/2020/1345054

CrossRef Full Text | Google Scholar

Yao, L., Deng, J., Johnston, R. J., Khan, I., and Zhao, M. (2018). Evaluating Willingness to Pay for the Temporal Distribution of Different Air Quality Improvements: Is China's Clean Air Target Adequate to Ensure Welfare Maximization. Can. J. Agric. Economics/Revue canadienne d'agroeconomie 67, 215–232. doi:10.1111/CJAG.12189

CrossRef Full Text | Google Scholar

Yi, Z.-F., Cannon, C. H., Chen, J., Ye, C.-X., Swetnam, R. D., Chen, J., et al. (2014). Developing Indicators of Economic Value and Biodiversity Loss for Rubber Plantations in Xishuangbanna, Southwest China: A Case Study from Menglun Township. Ecol. Indicators 36, 788–797. doi:10.1016/J.ECOLIND.2013.03.016

CrossRef Full Text | Google Scholar

Zhang, B., Yang, Y., and Bi, J. (2011). Tracking the Implementation of green Credit Policy in China: Top-Down Perspective and Bottom-Up Reform. J. Environ. Manage. 92, 1321–1327. doi:10.1016/j.jenvman.2010.12.019

CrossRef Full Text | Google Scholar

Zhang, J.-Q., Corlett, R. T., and Zhai, D. (2019). After the Rubber Boom: Good News and Bad News for Biodiversity in Xishuangbanna, Yunnan, China. Reg. Environ. Change 19, 1713–1724. doi:10.1007/s10113-019-01509-4

CrossRef Full Text | Google Scholar

Zhong, Zhen., Qi, Jie-li., Shi, Bing-qing., and Zhang, De-sheng. (2018). DoProfessional Farmers Have Higher Production Efficiency?——Evidencefrom Natural Rubber Growers in Yunnan and Hainan Provinces. Agric. Technol. economy, 40–51.

Google Scholar