1

Willingness to Pay for Cleaning Up Road Dust in Vientiane City

Phouphet KYOPHILAVONG Abstract This paper sets out to estimate residents’ Willingness to Pay (WTP) for cleaning up road dust in urban areas of Vientiane capital (VC) The analysis relies on the Contingent Valuation Method (CVM) and seeks to identify the factors’ affect on WTP. Based on 6,590 effective samples, empirical results show that mean WTP was high. The mean WTP is 7069 kip per person per month, which accounts for 4.71% of residents’ monthly income. In addition, education, income and have a statistically positive impact on WTP. Conversely, the number of children in family has a statistically negative impact on WTP. It is clear that residents may benefit significantly from cleaning up road dust in VC. These results offer crucial information to policy makers dealing with the dust problem on urban roads in VC. Keywords: Road dust, contingent valuation method (CVM), willingness to pay, and Vientiane capital. Introduction The adverse health effects of respiratory “dust” on human health are well-documented (Inyang and Bae, 2006). Recently, Vientiane capital (VC) faces the serious problem of road dust (figure 1). The dust contains MP 10, which has negative effects on the health of Vientiane residents and damages the city environment. Most of this dust stems from the increasing amount of construction, coupled with a lack of law enforcement to regulate illegal dust production. Until now, Vientiane capital authorities and the Water and Environment Authority (WREA) take few concrete actions to cope with the dust problems in VC. The government has neglected the dust problem for two main reasons: a lack of funds for cleaning up dust on roads, and a lack of understanding of costs of dust pollution for residents. Vientiane capital (CV) is capital city of Laos which has areas of 3920 Km2 with 667 thousand population (density of 171 person/Km2). VC has the fastest growing economy and population in Laos. Average economic growth in VC was more than 10 % and population growth was about 5 % over the past 5 years (VC, 2009). This growth has sped up public and private construction activities in city and it is one of the most important causes of dust on urban roads. The Lao government has undertaken some important policy and institutional measure to manage natural resource and environmental problems. Several

2

laws concerning natural resource and environmental issues have been passed in recent years. However, effective law enforcement has been hampered to environmental degradation. The main provision regarding environmental pollution (dust) is stipulated in the Lao PRD Constitution (1996) and Environmental Protection Law (1999). Unfortunately, enforcement of the law is weak. In addition, there are too many ministries and agencies involved in environmental management, producing inefficient responses to road dust problems. Moreover, the costs and benefits from cleaning up road dust are not well understood because there are no studies on the impact of dust on health and the environment in VC. Therefore, the main objective of this study is to estimate the willingness to pay (WTP) for cleaning up road dust in urban areas of VC and to identify the factors affecting their WTP. This study employed an open-ended Contingent Value Method (CVM) for estimating the willingness to pay (Hanemann, 1995; Alberini and Kahn, 2006; Takeuchi, 2000). The study also used multi-regression analysis to estimate the factors determining WTP. The scope of this study focuses on the impact of road dust on health problem and VC environment in urban areas of VC (see figure 2). Questionnaire design and survey The questionnaire was drafted based on some key sources (Alberini and Kahn,2006; Bateman and Will, 1999), interviews in government agencies and group discussions. Before the formal survey, two-pretests were conducted each with 20 samples. Based on the feedback from pre-tests, the questionnaire was revised accordingly. Interviews were limited to 20-30 minutes. The questionnaire consisted of four parts: (1) WTP on cleaning up road dust in urban Vientiane capital; (2) the Socio-economic characteristics of residents; (3) number of vehicle and main transportation used; (4) perception of the impact of dust on health and environment. In the willingness to pay (WTP) part of survey, the interviewee was asked the maximum WTP for reducing by 60-70% the health problems from dust problem and for reducing by 60%- 70% the environmental impact in urban area of Vientiane capital in 5 years though increasing water supply bill. The open-ended question was used for maximum WTP. Before asking this question, the negative impact of dust on health and environment were explained. Secondly, details of the project to clean up road dust in Vientiane city were explained. Thirdly, the impact of the project on reducing health risks and improving environment was described. Fourthly, the details of payment vehicle were also explained. In addition, pictures of current dust problems were also attached in the questionnaire and shown to interviewees during interviews. This survey was conduct from June to July 2009 in four districts of Vientiane capital. A total of 7625 samples were collected. This survey was carried out

3

by students and faculty lecturers from the Faculty of Economics and Business Management and the National University of Laos. Two sampling collection techniques were used and involved: face-to-face interviews and drop out interviews, and resident focused on head of household. The students and lecturers were trained before conducting the survey. Each of 4th students received 5-6 questionnaires from author; students then interviewed their neighbor. Socio-economic characteristics of respondents The socio-economic characteristics of the interviewee were investigated and are shown in table 1. About half of the respondents were female and about 30% were single and aged around 34 year old. On average, respondents had 13 years of education. About half were government officers. The average household size was four people, including one child. About 13 % of respondents had help problems, such as eye problems, sore throat, nose problems, lung / respiratory disease, heart disease, asthma, cold, and other diseases. Respondents had an average of two motorbikes and one car at home. More than 90% of respondents considered dust to be a problem for their health and the environment of the city.

4

Table 1. Socio-economic characteristics of respondents Socio-economic condition

Female (%) 47.0Single (%) 29.4Age (year) 34.8Education (year) 13.5Income (10000 kip)/month 15.4Home locates near main road (%) 30.0Main occupation (%)Government 47.6Private 18.7Retails/owner of business 20.6Student 9.8Others (%) 3.3Family and children (number)Family 4.7Children 0.7Having problem/disease (%)*Eye 21.1Throat 18.4Breath 12.9Nose 8.0Cool 27.0Fever 9.6Heart 3.4Other disease 6.1Having vehicle (number)Bicycle 0.5Motorbike 2.1Car 0.7Perception on the impact of dust (%)Serious impact on health 96.8Serious impact on city environment 92.9Note: Due to multiple choices question, the total ratio is more than 100% Willingness to pay A total of 7625 questionnaires were collected.6,590 of these questionnaires (86 % of total samples) contain effective information. In 6,590 samples, 4951 interviewee (24.9%) show zero WTP, and 4951 interviewee (75.1%) show positive WTP. The whole sample (6,590 samples) of mean WTP to 60-70% reducing health problems from dust problem and 60%- 70% improving environment was 7069 kip or 0.86 US$ per household per month. This accounted for 4.71% of income per month (table 2). Ashown in table 3,

5

WTP of 1000-5000 kip/person/month is the highest share, which covered 37.7%.WTP of 6000-10000 kip/person/month is the second highest share, accounting for 15.3 % (table 4). There are 369233 persons living in the urban area of VC (appendix 1).Therefore, the aggregate WTP of residents is 317.5 thousand US$ per month. In addition if only with positive WTP (4951 sample) is considered, the mean WTP is 9410 or 1.15US$ per household per month, which accounts for 5.80% of income per month. The aggregate WTP of residents is 424.6 thousand US$ per month. The main reason for paying to cleaning up the road dust condition is for the cleanness of the city (45.5%), to pay for personal health and that of family members (31.9%) and to contribute to government (13.0%) (table 5). There are some gaps in applying CVM in Laos for to two main reasons; Firstly Laos is a socialist country, which has been moving towards a more open market. Secondly, public goods are responsibility of government and households has little experience in public consultation and problem solving involving. Open-ended questions might give a lower mean WTP compared with other types of questionnaire (Hausman, 1993). The estimation of WTP in developing countries is also relatively low (Whittington, 2002).However, by comparing the mean WTP to studies in other developing countries, the mean WTP of this study is high. For example, the mean of WTP per household per year for 50% reduction of harmful substance in the air in Beijing was 0.7% of annual income (Wang et al., 2006). Some bias for high mean WTP may, however, be apparent. Firstly, respondents may not consider the time horizon of the payment. In this survey, the time horizon of the payment is 5 years. Secondly, respondents may not distinguish between personal and household payments. This survey focuses on individual payment. Thirdly, as in many developing countries, the salary of government in Laos is low and they have other sources of income to support their family. Respondents might not reveal all income they received. However, the author believes that the road dust problem is a serious problem for them, therefore their mean WTP is high. Open-ended questionnaire have been criticized (Arrow et al., 1993; Hanemann,1994). This type of questionnaire, however, still has the advantages of the effective utilization of the data and enables an avoidance of the starting point and yes-saying bias (Carlsson and Johansson-Stenman, 2000). However, there are no studies concerning with residents’ WTP for improving air quality in Laos, therefore, it is difficult to design a bid-vector for discrete choice questions or a scale and intervals for payment card questions.

6

Table 2. Mean WTP of residents (kip)

WTP/IncomeMean STDEV Median Mode Minimum Maximum (%)

Whole sample (N=6590)7069 11432 5000 0 0 150000 4.71

Posive WTP (N=4951)9410 12328 5000 5000 200 150000 5.80

WTP

Table 3. Range of WTP of residents

Willingness to pay (kip)

Number of interviewees

Percentage (%)

0 1639 24.9<1000 478 7.31000-5000 2482 37.76000-10000 1006 15.311000-15000 267 4.116000-20000 364 5.521000-25000 59 0.926000-30000 107 1.631000-35000 16 0.236000-40000 19 0.341000-45000 8 0.146000-50000 97 1.551000-55000 3 0.056000> 45 0.7Total 6590 100.0 Table 5. The reasons for positive WTP

Reason for payment PercentCleanness of city 45.5Health of yourself and family member 31.9Contribution to government 13.0High indirect benefit from eliminate dust 6.7Other reasons 2.9 On the other hand, the main reasons of zero WTP are as follows. 42.6 % of zero WTP’s respondents said that dust was caused by trucks and construction activities; 28.3% of respondents said that it was the government’s duty to clean up road dust in urban areas of Vientiane capital; while14% of residents mentioned that they did not have confidence in using their money to implement the project.

7

The zero WTP could have two categories: true zero WTP and protest bids (Wang et al., 2006). For the protest bids, respondents might have difficulty understanding the valuation questionnaire. It is difficult to separate the zero WTP from true condition and protest bids. Therefore, I did not exclude any of the responses as protest bids for analysis. Table 6. The reasons for zero WTP

Reasons not to pay PercentIt is government's duty, not residence 28.3It is responsibility of producer (trucks, construction activities) 42.6Do not have money, other expenditure is high 10.5Do not have confidence of using money and implementation of this project 14.2Dust in street is not big problem 1.3Other reasons 3.1 Comparing characteristics of respondents of zero and positive WTP I used a simple t-test to investigate the socio-economic effects on zero WTP and positive WTP of households (table 7).There are statistically significant differences between two groups in term of sex, year of education, main occupation (Government, retails shop owner/ business owner, and student), income, proximity of home to the road, pre-existing health problems (number of illness), number of cars at home and perception on the impact of dust on health and environment. Here, it is clear that positive WTP residents have higher income, education, lower number of illnesses, and higher perception of the detrimental effects of road dust on health and environment than zero WTP residents.

8

Table 7. T-test for socio-economic variables of zero and positive WTP

WTP=0 WTP>0 Different WTP=0 WTP>0Sex 0.49 0.46 0.02 0.50 0.50 1.74 0.08Age 35.08 34.76 0.32 10.08 9.88 1.13 0.26Status 0.30 0.29 0.01 0.46 0.46 0.41 0.69Education 13.23 13.61 -0.39 2.97 3.20 -4.47 0.00Main occupationGovernment 0.50 0.47 0.03 0.50 0.50 2.03 0.04Private company 0.19 0.19 0.00 0.39 0.39 -0.05 0.96Retails shop owner/business owner

0.18 0.21 -0.03 0.38 0.41 -3.12 0.00

Student 0.02 0.04 -0.02 0.14 0.19 -3.94 0.00Income 12.99 16.23 -3.24 15.43 23.99 -6.33 0.00Home locates near main road 0.28 0.31 -0.03 0.45 0.46 -2.30 0.02Number of person at homeFamily member 4.78 4.71 0.07 2.10 1.99 1.22 0.22Children member 0.64 0.67 -0.03 0.84 0.88 -1.05 0.29Having health problems (number of illness)

0.96 1.10 -0.14 1.08 1.10 -4.50 0.00

Number of car at homeBicycle 0.50 0.53 -0.02 0.86 0.86 -0.95 0.34Motorbike 2.06 2.05 0.01 1.19 1.19 0.20 0.84Car 0.68 0.71 -0.03 0.88 0.89 -1.34 0.18Perception on Impact of dust on health 1.48 1.41 0.07 0.66 0.64 3.49 0.00Impact of dust on environment 1.59 1.55 0.04 0.82 0.80 1.94 0.05

Socio-economic characters Mean STDEV t-value Significance

The factors affect WTP In order to evaluate the influence of socio-economic characteristics on WTP, the multiple regression model was used. The definitions of variables in WTP function is shown in table 8. I used the OLS (Ordinary Least Square) method to estimate the factors affect the WTP. The multicollinearity in the independent variables has checked by correlation matrix method. The variables which had correlation of less than 50% to other variables were chosen to use in regression. The result of the factors affect WTP is shown in table 9. As expected, income has a statistically significant impact on WTP. In addition, some socio-economic characteristic variables has also had an affect on WTP. And other socio-economic variables: marital status, education, retail shop owner/business owner, house locating near main road, number of family members, having other diseases, and number of bicycle, motorbike, and car at home have positive significant impacts on residents’ WTP. Here, it indicates that residents who have high income and high education have high influence on WTP. This result is consistent with other studies (Wang, 2006). On the other hand, three socio-economic variables; gender, number of children in family, and perception on the impact of dust on the environment have negative significant impacts on residents’ WTP. This indicates that

9

residents who are female, single, and have a large number of children have a negative impact on WTP. Table 8. Definition of variable in regression Variables Definition Unit Mean Std. Dev. Minimum Maximum

Expected sign

Independent variableWTP Willingness to pay kip 9408.3 12322.7 200 150000 NADependent variablesGEND Gender Female=1, other=0 0.5 0.5 0 1 ?AGEY Number of year Year 34.8 9.9 14 76 +STAT Statute of marries Single=1, other=0 0.3 0.5 0 1 +EDUC Year of attending school Year 13.6 3.2 0 18 +Main occupation GOVE Government officer yes=1, other=0 0.5 0.5 0 1 +RETA Shop owner/business owner yes=1, other=0 0.2 0.4 0 1 +STUD Student yes=1, other=0 0.0 0.2 0 1 +INCO Income per month 10000 kip 16.2 24.0 1 535 +MARO House locates near main road yes=1, other=0 0.3 0.5 0 1 +Number of person and child at homeFMUM Family member Person 4.7 2.0 1 13 +CHIL Child Person 0.7 0.9 0 7 -Health situationEYES Eye problems yes=1, no=0 0.2 0.4 0 1 +THRO Sore throat yes=1, no=0 0.2 0.4 0 1 +BREA Lung / respiratory disease yes=1, no=0 0.1 0.3 0 1 +NOSE Nose problems yes=1, no=0 0.1 0.3 0 1 +COLD Cold yes=1, no=0 0.3 0.5 0 1 +ASTH Asthma yes=1, no=0 0.1 0.3 0 1 +HEAR Heart disease yes=1, no=0 0.0 0.2 0 1 +ODIS Other disease yes=1, no=0 0.1 0.2 0 1 +Number of vehicle at homeCARH Car Number 0.3 0.4 0 1 +BICLE Bicycle Number 0.0 0.1 0 1 +MOTO Motorbike Number 0.7 0.5 0 1 +Perception of serious dust problem on HEAL Health yes=1, no=0 1.0 0.2 0 1 +ENVE City environment yes=1, no=0 0.9 0.3 0 1 +

10

Table 9. Factors affect WTP Variables Coefficient t-statistic P-value

GEND -679.7 -1.87 0.06AGEY -10.1 -0.45 0.66STAT 1094.1 2.27 0.02EDUC 130.9 2.04 0.04Main occupation GOVE 78.1 0.18 0.86RETA 1008.3 1.94 0.05STUD 1636.9 1.60 0.11INCO 25.6 3.31 0.00MARO 720.5 1.88 0.06Number of person and child at homeFMUM 328.3 3.35 0.00CHIL -492.8 -2.24 0.03Health situationEYES -667.9 -1.49 0.14THRO -350.6 -0.77 0.44BREA 25.3 0.05 0.96NOSE -232.2 -0.35 0.73COLD -398.7 -0.99 0.32ASTH 792.3 1.31 0.19HEAR 776.0 0.81 0.42ODIS 1957.3 2.17 0.03Number of vehicle at homeCARH 3418.7 3.13 0.00BICLE 5113.2 2.64 0.01MOTO 2176.5 2.07 0.04Perception of serious dust problem on HEAL -1068.1 -0.87 0.38ENVE -3482.5 -3.36 0.00Constant 7749.0 4.39 0.00Number of observationAdjusted R2

F-statistic

49440.01784.74

11

Conclusion The main objective of this paper is to estimate residents’ Willingness to Pay (WTP) for the clean up of road dust in urban areas of Vientiane capital (VC) of Laos using the Contingent Valuation Method (CVM) and to identify the factors’ affect on WTP. The WTP for cleaning up road dust in urban areas of VC is high.60-70%, reducing health problems from dust problems and 60%- 70% improving environment was 7069 kip or 0.86 US$ per household per year or account for 4.71% of household income per month. The main reason for this payment is for the cleanness of the city, improvement in health conditions and contribution to government. Residents who have higher income and higher education have higher WTP. Therefore, I can conclude that the residents’ WTP to clean up road dust in VC is high. In other words, resident will benefit from cleaning up road dust in Vientiane capital. It is crucial information for government agencies to consider more seriously and it will help them understand the on costs and benefits of cleaning up road dust and taking effective actions to enhance law enforcement to stop illegal causes of road dust in Vientiane capital. In addition, there is very few studies on valuation of public goods in Laos. Therefore this study can be part of the important step towards the uses of CVM to evaluate non-market goods for Laos in the near future. However, this analysis faces several weaknesses. Firstly, this study may face some bias. Secondly, it used easy data sampling collection methods, which might be insufficient in themselves for sampling data. It is important to use stratified random sampling techniques. Thirdly, this study used simple CVM- open-ended questionnaires, which might not be an effective method. Dichotomous choice CVM or Choice Modeling might be appropriate for future studies. Acknowledgements I would like express my sincere appreciation to students, lecturers and my research assistants at the Faculty of Economics and Business Management, National University of Laos for helping me collect data, and the Deputy Director of the Department of Environment and the Water Resources and Environment Authority (WREA) for their comments and cooperation. The author alone is responsible for any errors in this report.

12

References Alberini, A., and Kahn, J. R. (2006). Handbook on Contingent Valuation, UK:

Edward Elgar. Arrow, K., Solow, R., Portney, P. R., Leamer, E. E., Radner, R. and Schuman,

H. (1993).Report of NOAA panel on contingent valuation. 58 Federal Register 4601 (January 15).

Bateman, I.J., and Will, K. eds. (1999), Valuing Environmental Preferences: Theory and Pracitces of the Contingent Valuation Method in US, EU and Developing Countries, Oxford University Press.

Carlsson, F., and Johansson-Stenman, O. (2000). Willingness to Pay for Improved Air Quality in Sweden, Applied Economics, Vol. 32, Issue 6, pp. pp. 661-669.

Hanemann, W. M. (1994) “Valuing the environment through contingent valuation”, Journal of Economic Perspectives 8 (4), 19-43.

Hausman, M. (1993). Contingent Valuation: A Critical Assessment, North- Hollend, Amsterdam.

Inyang, H. I and Bae, S. (2006), Impacts of dust on environmental systems and human health, Journal of Hazardous Material 132, pp.6-7.

Takeuchi, K. (2000), Contingent valuation method and travel cost method, Japan: Keisoshobo.

Kahneman, D. and Knetch, J. L. (1992).Valuing Public Goods: the Purchase of Moral Satisfactions, Journal of Environmental Economics and Management, Vol 22, Issue 1, pp. 57-70.

VC (Vientiane capital). (2009), Socio-Economic Statistics of Vientiane Capital, Division of Planning and Investment, Vientiane Capital, Vientiane, Laos.

Wago, H., and Ban, K. (1995). Applied Econometric Analysis Using TSP 2rd, University of Tokyo Press, Tokyo, Japan.

Wang, X.J., Zhang, W., Li, Y., Yang, K.Z., and Bai, M. (2006). Air Quality Improvement Estimation and Assessment Using Contingent Valuation Method, A Case Study in Beijing, Environmental Monitoring and Assessment, Vol. 120, pp. 153-168.

Whittington,D. (2002). Improving the Performance of Contingent Valuation Studies in Developing Countries, Environmental and Resource Economics, Vol 22, No. 1-2, pp. 323-367.

13

Figure 1. Road dust in urban areas of Vientiane capital

Note: these pictures were used in questionnaire Figure 2. Central districts in Vientiane city

14

Appendix1. Population in survey areas

District Nol Villages Household Population(Person)

Chanthabuly 37 11778 75902

Sikhothottabong 61 19313 110129

Xaysethha 52 17093 107489

Sisattanak 40 10491 75713

Total 190 58675 369233

Source: VC (2009).