Using choice experiment, the objective of this study is to estimate international tourists’ preferences and willingness to pay for ecofriendly services which related to river and natural environment conservation of hotel industry in Vangvieng, a popular tourist town in Lao PDR. We used the mixed logit to take into account of preference heterogeneity by allowing coefficients to be normally distributed and assumed to vary among individuals. The result shows that all ecofriendly practices offered in this study are significantly preferred by tourists. Income, age, and education did not have significant effect on the probability of choosing ecologically friendly practices of hotel and attributes to support the ecotourism of the area; however, this study provides important information that female tourists are more receptive to a new alternative. These findings do not only support the provision of good and green services in Vangvieng Town, but will also be useful for policy maker to enhance the sustainability of ecotourism in Laos.
Rivers play a key role in ecosystems and society, and they provide a range of ecosystem functions such as shelter and food source for an array of biological species, aid in flood management and ecological refuge development [
As a land locked, but rich with natural and fresh water sources among the Lower Mekong Basin countries, Lao PDR has great potential for tourism industrial development, especially for natural resource and river based- ecotourism [
Covering 180,434 hectares of river basin with around 106 Km length and flowing through larger areas of three districts, the Nam Xong River is not only important for local livelihood but the most economically beautiful and unique river for tourism in Vientiane province. Especially in Vangvieng district, where most of tourist attractions and activities are based on natural resources, Namxong River in particular [
Nevertheless, some challenges for sustainable tourism development are still concerned. The rapid growth of tourism industry is expected to exert an ever great pressure on the environment, society and culture. The lack of direct participation of communities, financial constraints and limited access to credit for conservation are the key challenges facing sustainable ecotourism development [
In Vangvieng, it has been disputed that rapid urban development is detrimental on natural environment, and remedial action is required [
In spite of the partnership between national protected area (NPA) managers, local communities and private sectors in provision of good and green services in tourist sites has been promoted [
On the other views, while at least 29 green attributes mentioned by Environmentally Friendly Hotels [
While water rate or one-off payment are often found in studies on evaluation of non-marketed values of rivers through households’ willingness to pay (WTP) and preferences [
To estimate economic value of this Nam Xong River conservation, 3 green attributes which closely link to river conservation are used in this study, including water quality conservation, local handicraft products in guestroom and responsibility for biodiversity conservation along the river. By using choice experiment, the objective of this study is to find out which of these ecofriendly services are preferred by international tourists. The finding is to provide more appropriate information about market demand which is important for policy maker to improve service quality of hotel industry and conservation programs in this town.
CE is one of choice modeling technique that close links with economic theory which allow the results to be interpreted as being equal to marginal (or total) values for use in CBA [
According to Hensher et al., 2005, Vij is denoted as a deterministic utility that can be explained by the attributes in alternative j in the relevant choice set as well as by the respondent i’s characteristics. ɛij is the random error term that indicates the unknown factors about the respondent i or imperfect information that cannot be explained by attributes in alternative j in the choice set. The indirect utility (Uij) can be expressed:
Assuming that in the relevant choice set consists of J alternatives
then
where, disturbance term (ɛ) is assumed to be independently and identically distributed (IID) with an extreme-value type 1 (EV1) distribution. The probability of the respondent i choosing the alternative j out of the set of J alternatives equals to the ratio of the exponential of the observed utility index for alternative j to the sum of the exponential of the observed utility indices for all J alternatives, including the jth alternative. The conditional logit (CL) model known as a specification of general multinomial logit (MNL) model can be estimated as follow:
The log-likelihood function is expressed in the Equation (6), where yij is an indicator variable which takes a value of 1 if respondent i choose alternative j and zero otherwise:
However, the MNL is most commonly used in discrete choice model for the analysis of results from CEs and its relative simplicity is a clear advantage. Consequently, the restricted condition that the homogenous preference across respondents and the independence of irrelevant alternative (IIA) assumptions needs to be satisfied always entails the MNL facing some important limitation [
The utility associated with each alternative j, as evaluated by each individual i in choice situation t, is represented in a discrete choice model by a utility expression of the general form:
where Xij is the full vector of explanatory variables that can be observed, including attributes of the alternatives, socio-economic characteristics of the individual and descriptors of the decision context and choice task itself in choice situation t. βi and
The unconditional choice probability is the expected values of the logit probability over all the possible values of βi, that is, integrated over these values, weighted by the density of βi
Unlike MNL, the simulation technique of sequences needs to be applied to calculate the probability in the mixed logit model [
The main purpose of this study is to find out an appropriate way to encourage private sector, especially hotel industry in this Vangvieng Town, to participate in sustainable natural environment conservation program by providing their guests with ecofriendly services that related to river conservation within their hotel while gaining more benefits. But a question arises to which ecofriendly or green practices are most preferred among several green attributes. According to the rapid growth in number of hotel with various sizes and prices along the river and the discussion with the natural resources and environment office of the district during pilot survey in September 2014, basic green levels of practice, such as wastewater recycling system, still strongly needs to be taken into account. Based on green attributes defined by Environmentally Friendly Hotels, 2014 [
Totally, 54 combinations are produced from 3 attributes each with 3 levels and 1 attribute with 2 levels (33 × 21). By using SPSS software program [
In this study, it is alternatively hypothesized that these ecofriendly attributes indifferently influence on the probability that tourist will choose for new service offered by hotel. Also Main socio-economic factors affect indifferently on the probability of choosing new alternative offered in choice set.
Attributes | Level |
---|---|
Water quality conservation(WQC) | 1) No (No conservation), 2) Wastewater recycling only (WRO), 3) Wastewater recycling + organic soap & amenities (WROSA). |
Cotton towel, bed sheets and furniture in guestroom (CTBF) | 1) Local handicraft products (LHP), 2) Non-local handicraft products (NLHP). |
Biodiversity conservation along the river (BIOC) | 1) 0 ha (No conservation), 2) 10 ha, and 3) 20 ha. |
Additional price/night (AP) | $0, $1, $3 and $5. |
Attributes | Option A | Option B | Option C |
---|---|---|---|
Water quality conservation | Wastewater recycling only | Wastewater recycling + Organic soap set | I would not choose any of these options I prefer the same service |
Cotton towel, bed sheets & furniture | Non-local handicraft products | Local handicraft products | |
Biodiversity conservation | 0 ha (No conservation) | 10 ha along the river | |
Additional price/night | $1 | $3 | |
I would choose. | ☐1 | ☐2 | ☐3 |
Face to face interview was conducted in Vangvieng Town during November, 2014. Totally, 200 international tourists from 27 countries were randomly interviewed.
Each questionnaire was divided into 4 parts. Part one, respondents were asked some introduction questions about sources of information about this tourist town, frequency of visit, inspiration to visit the town, hotel room price and length of stay; then followed by questions about agreement with the role of green products in tourist site and experiences of staying in an ecofriendly hotel. Part two, respondent was explained about each ecofriendly service that will be offered by hotel and also its relation to conservation of the river and natural environment before being asked to express their support in this part and preference by choosing the most preferred alternative offered in each choice set in part three. Finally, the information about social demographic of respondents was asked in part four.
The largest respondents were from Europe and Asia 47% and 37% (
On average, respondents spent around 3 days or 2 nights visiting the town. While average value of hotel price mentioned by respondents was $14/night, the cheapest and the most expensive prices were ranged only from $3/nigh to $85/night. Among 200 respondents only 10 of them did not complete or not answer choice questions. Therefore, 190 complete questionnaires with 1008 observations out of 3025 choice sets were used in the analysis. In this study, all variables are dummies, except only the price which is a continuous variable (
Category | Frequency | Percent | |
---|---|---|---|
Region | Europe Asia Other | 94 74 32 | 47 37 26 |
Age group Gender Marital Status | 18 - 25 26 - 30 31 - 40 41 - 50 51 - 60 >60 Male Female Married single other | 67 64 38 12 12 7 95 105 48 136 16 | 34 32 19 6 6 4 48 52 24 68 8 |
Education level | ≤High school Vocational or collage ≥University | 27 29 144 | 13.5 14.5 72 |
Occupation | Government Private Student Other | 22 130 33 15 | 11 65 16.5 7.5 |
Income group | <$2000 2000 - 5000 6000 - 10,000 11,000 - 15,000 16,000 - 20,000 21,000 - 30,000 31,000 - 50,000 51,000 - 70,000 71,000 - 100,000 >$100,000 | 21 16 20 27 5 37 28 30 12 4 | 10.5 8 10 13.5 2.5 18.5 14 15 6 2 |
Data source: field survey 2014, November.
Variable | Description and codification | Mean | Std | Min | Max |
---|---|---|---|---|---|
ASC | Alternative Specific Constant (ASC = 1 if one of new alternatives is chosen and 0 = otherwise) | ||||
WRO | Water quality conservation by using wastewater recycling system only (WRO = 1 if WRO is chosen and 0 = otherwise) | 0.24 | 0.43 | 0.00 | 1.00 |
WROSA | Water quality conservation by using wastewater recycling system and serving with organic soaps and amenities | 0.17 | 0.38 | 0.00 | 1.00 |
LHP | (WROSA = 1 if WROSA is chosen and 0 = otherwise) Cotton towels, bed sheets and furniture in guestroom made from local communities (LHP = 1 if LHP is chosen and 0 = otherwise) | 0.34 | 0.47 | 0.00 | 1.00 |
BIOC1 | Biodiversity conservation along the river within 10 ha (BIOC1 = 1 if 10ha conservation is chosen and 0 = otherwise) | 0.25 | 0.43 | 0.00 | 1.00 |
BIOC2 | Biodiversity conservation along the river within 20 ha (BIOC2 = 1 if 20 ha conservation is chosen and 0 = otherwise) | 0.69 | 0.37 | 0.00 | 1.00 |
AP | Additional price for hotel room per night and takes the value $0, $1, $3 and $5. | 1.58 | 1.93 | 0.00 | 5.00 |
Non-attribute variable | |||||
MALE EUT | Gender of respondent (MALE = 1 if male and 0 = female) European tourist (EUT = 1 if European tourist, 0 = otherwise) | 0.45 0.47 | 0.50 0.50 | 0.00 0.00 | 1.00 1.00 |
According to the Equation (1), the observable component of the indirect utility function can be expressed by the sum of attributes’ part-worth utilities of the respondents [
where, β1 represents the coefficient of the alternative specific constant and β2, β3, β4, β5 and β6 denote the coefficients of attributes associated with the ecofriendly services provided by hotel that related to river conservation.
Using the econometric software Nlogit Version 5 and under the assumption that coefficients across individual are constant and the choice sets comply with the IIA condition [
The positively and highly significant coefficient of LHP suggests that the respondents more prefer the cotton towels, bed sheets and furniture made from local communities to those that are non-local products. This can be
Variable | Homogeneity | Heterogeneity | ||
---|---|---|---|---|
MNL | Extended MNL | Mixed Logit | Extended Mixed Logit | |
ASC WRO WROSA LHP BIOC1 BIOC2 AP | −0.159 (0.197) 1.073*** (0.127) 1.621*** (0.144) 0.984*** (0.118) 1.650*** (0.138) 1.796*** (0.149) −0.146*** (0.028) | 0.176 (0.256) 0.479*** (0.161) 1.352*** (0.180) 1.008*** (0.121) 1.645*** (0.139) 1.827*** (0.153) −0.157*** (0.050) | 1.759* (1.048) 1.756*** (0.302) 2.603*** (0.417) 2.191*** (0.435) 2.670*** (0.359) 2.794*** (0.435) −0.358*** (0.082) | 4.210** (2.142) 0.717** (0.333) 2.394*** (0.548) 2.556*** (0.662) 2.953*** (0.580) 3.188*** (0.731) −0.452*** (0.129) |
Non-attribute variables | ||||
ASC*MALE WRO*EUT WROSA*EUT Std.Devs WRO WROSA LHP BIOC1 BIOC2 | −0.566** (0.269) 1.372*** (0.250) 0.693** (0.270) | 0.159 (1.062) 0.802 (0.705) 3.264*** (0.666) 0.667 (0.526) 0.500 (0.962) | −1.757* (0.947) 2.994*** (0.898) 1.617** (0.662) 0.041 (0.810) 1.167 (0.808) 3.874*** (1.041) 0.724 (0.748) 0.962 (0.947) | |
Log likelihood Pseudo R2 (ρ2) | −650 0.260 | −629 0.280 | −627 0.434 | −602 0.456 |
Note: ***, **, *Significance at 1%, 5%, 10% level.
mentioned that the international tourists would also like to support local people income from ecotourism while they can enjoy traditional local products and furniture in their hotel rooms.
With highly significant level, the greatest and nonnegative coefficient of both BIOC1 and BIOC2 indicates that the probability of choosing new alternative will increase if the area that hotel takes responsibility for biodiversity conservation increases. However, the probability that a new alternative with BIOC2 will be chosen is greater, there is no completely big difference in value between the coefficients of BIOC2 and BIOC1. As a result, it can be observed that the probability that although a new alternative with 20 ha biodiversity conservation is more likely to be chosen over those with 10 ha; the propensity of increasing in probability is dramatically reduced when the area of conservation moving from 10 to 20 ha. This implies that the area of conservation should be limited in an appropriate scale.
As regards, with interactions MALE in the alternative specific constant and EUT in the attributes WRO and WROSA, the extended MNL in column 2 obtains statistically better than model 1. According to the result of the significant interaction terms, it indicates that explanatory power of model is improved [
The negative coefficient of MALE variable interacted in ASC implies that female tourist’s probability of choosing for new alternative is higher than male tourist’s. The positive coefficients of WRO*EUT and WROSA*EUT indicates that the probability that new alternatives will be chosen by European tourist will increase if WRO or WROSA is not excluded from the alternatives. This implies that European tourists are more willing to contribute the efficiency of water use of hotel industry than tourists from other region.
Using Hausman test for IIA by excluding alternative 1 from choice sets [
Like the extended MNL, interacting MALE in the ASC and EUT in WRO and WROSA, the result of extended RPL in column 4 shows slightly lower coefficient of WROSA and the same significant level at 1%. But while the coefficient of WRO is lower in magnitude and significant at only 5%, the rest has greater coefficients and the same 1% level of significance. The coefficient of the ASC is also positive and significant at 5%. Again according to highly significant at 1% level of log-likelihood ratio test against the RPL model and pseudo R- square (0.456), the result of extended RPL model can be statistically superior and more informative. Nevertheless, even the signs of coefficients of all variables remain unchanged; their ranking is different. The coefficient of BIOC2 is still largest and followed by BIOC1and LHP, while WROSA is in the fourth place and lowest for WRO.
In term of interaction, the result of the extended RPL model shows that sign of all interactions’ coefficient are still the same as the extended MNL and also the level of significant at 5% and 1% for two interactions WRO*EUT and WROSA*EUT. However, lower significant level is found for ASC*MALE at 10%, these interactions still provide the same information as in extended MNL. Similarly, both coefficient and standard error of all attributes and interaction terms are absolutely changed upward. This confirms the heterogeneous preferences of the international tourists towards the ecologically friendly practices of hotel industry in this Vangvieng Town.
The marginal willingness to pay (MWTP) for one ecofriendly service attribute can be calculated by dividing estimated coefficient on the ecofriendly service attribute of interest by the negative coefficient on the additional price attribute [
In
The results from the ASC gives the information that without any of these offered green attributes, the respondents have unobservable willingness to pay additional price to support the hotel industry up to $7.54/night. This can be mentioned that international tourists need hoteliers to participate in carrying of an environment program. In addition, the results from interactions show that European tourists are willing to pay higher additional price for WROSA and RWO than tourists from other regions.
The above mentioned implicit prices do not provide estimates of compensating surplus (CS) for new ecofriendly service alternative scenarios, but welfare measurement can be derived from the marginal rate of substitution between the residual of the initial and alternative utility states divided by the marginal utility of income, which represented by coefficient of “price” attribute [
In order to estimate the social benefits, the CS can be estimated under the status quo and 2 new ecofriendly service scenarios below:
・ Current scenario (status quo): No water quality conservation (no neither only wastewater recycling nor wastewater recycling with organic soaps and amenities), no local handicraft products in guestroom and no biodiversity conservation along the river.
・ Scenario 1: Water quality conservation using WRO, local handicraft products in guestroom (LHP), 10 ha biodiversity conservation along the river (BIOC1).
・ Scenario 2: Water quality conservation using WROSA, local handicraft products in guestroom (LHP) and 10 ha biodiversity conservation along the river (BIOC1).
Variable | Extended RPL | |||
---|---|---|---|---|
Base | EUT | MALE | Weighted MWTP | |
ASC WRO WROSA LHP BIOC1 BIOC2 | 9.30** 1.59* 5.29*** 5.65*** 6.53*** 7.05*** | 8.20*** 8.86*** | 5.42** | 7.54** 4.71*** 6.98*** 5.65*** 6.53*** 7.05*** |
Note: ***, **, *Significance at 1%, 5%, 10% level.
・ Scenario 3: Water quality conservation using WROSA, local handicraft products in guestroom (LHP) and 20 ha biodiversity conservation along the river (BIOC2).
To estimate overall WTP, it is necessary to include ASC, which captures the systematic but unobserved information about respondent’s choices [
As regards, the CS for the change from the status quo to the scenarios increases as the levels of ecologically friendly practice of hotel is improved. In
To enhance the sustainable development in ecotourism, the participation of all stakeholders in provision of ecofriendly products and services in tourist sites as well as in tourist town is the most important factor in one hand. On the other hand, customer’s behavior and preferences for those green products or services is also a crucial and usually defines what kind of the products should be provided and how much the price should be imposed.
This study on economic evaluation of river conservation towards tourist’s preference for ecofriendly services of hotel industry which related to river conservation by using choice experiment valuation technique, the findings from both MNL model and mixed logit model as well as the extended ones show expectedly that all ecofriendly service attributes offered in the survey are significantly preferred by international tourists. Green practice that conserves biodiversity and natural forest along the river within 20 ha or 10 ha are all preferred with the highest and high MWTP for additional room price. This result evidently suggests that the hotel industry in this town should express their responsibility for biodiversity conservation along the river for the sake of their customers’ welfare while providing more benefit for environment and gaining more benefit as well as stabilizing their own businesses in the short and long run.
Water quality conservation practice by using wastewater recycling system and serving with organic soaps and amenities or WROSA is more preferred to the practice that only wastewater recycling system or WRO with the difference of MWTP up to $2.27. So WROSA is suggested to provide. In addition, the interaction result indicates that European tourists have higher concern on the impact of wastewater from hotel industry on the natural environment in this tourist town while wastewater recycling system can also indeed reduce high level of freshwater demands of hotel industry [
With high value of MWTP, another ecofriendly attribute which should be considered and provided is the service serving with cotton towels, cotton bed sheets and handicraft furniture in guestroom which are handicraft products made by local communities. The preferences towards this attribute of international tourists also links to their willingness to support local communities’ income from ecotourism while the can enjoy locally traditional products in their hotel rooms. Also, increasing in benefit from ecotourism will consequently increase the attitudes towards environment conservation of local communities as well [
However, the effects of main social-economic demographic factors (income, age and education) which are usually found in many studies on choice are not significantly found. The result of this study provides important information that regardless of income, age and education, those who visit the town are willing to pay for ecologically friendly practices and support the ecotourism of the area while enjoying their stay. Significantly, the finding from gender factor gives a conclusion that female tourists are more receptive to new ecofriendly services. Therefore, appropriate promotion should be considered to increase female tourists.
Scenario | Extended RPL Model |
---|---|
Scenario 1 Scenario 2 Scenario 3 | 17.90*** 26.69*** 27.21*** |
Note: ***, **, *Significance at 1%, 5%, 10% level.
In addition to providing more detail of market information for sustainable conservation of the Nam Xong River and ecotourism development in this town, preference of domestic tourists and other hotel’s customers as well as the participation of hoteliers in provision of good and green services should be properly conducted.
PhouthivongKhamsay,YoshifumiTakahashi,HisakoNomura,MitsuyasuYabe, (2015) Economic Valuation of River Conservation towards International Tourists’ Preferences and Willingness to Pay for Ecofriendly Services of Hotel Industry: A Case Study of Namxong River in Vangvieng District, Laos. Journal of Water Resource and Protection,07,897-908. doi: 10.4236/jwarp.2015.712074