| Scientific evaluation of the ecosystems' economic value is the basis of promoting environmental protection and economic cooperation,which is helpful for rational development and utilization of environmental resources and scientific formulation of environmental protection policies.As a key method of non-market value assessment,Contingent Valuation Method(CVM)has been widely used in research fields,such as traffic management,food,environmental economics.In recent years,CVM's evaluation results have been controversial due to its hypothetical characteristics.The controversy centers on the widespread phenomenon of respondents' refusal to pay during CVM's survey.The derivative problem of zero payment and its treatment have become the focus of CVM's research.The treatment of previous CVM evaluation models of zero-payment samples are prone to sample selection bias and estimation bias,which fail to effectively solve the impact of zero-payment on evaluation results.To solve this problem,according to the characteristics and distribution of respondents' refusal to pay,this paper divides the zero-payment samples into two categories: one is the true zero willingness to pay,the marginal utility of consumers is not affected by the advantages and disadvantages of research objects;the other is the protest willingness to pay,the survey mode,cognitive level and other factors make the respondents hide their real needs for research objects.Binary logit model was used to identify two types of determinant variables of zero payment and analyze the internal mechanism of sample payment difference.Based on the random utility theory,zero payment samples are included in the WTP statistical analysis,and the spike model adapted to double bounded dichotomous CVM data is built to analyze and compare the differences between the spike model and the random utility model.At the end of the paper,based on the cross-case perspective,the monetization evaluation of ecological protection of the Sanjiang Plain wetland and Zhalong wetland in Heilongjiang Province is conducted to demonstrate the stability of the spike model in CVM research.The results show that:(1)income and government trust are the main reasons for respondents' attitude of zero-payments.Unlike “income restriction”,government trust issue is the main determinant variable of the protest against high or low payment rates.They are prevalent in CVM studies in developing countries.Distrust discourages public participation in government-sponsored and implemented projects;(2)the average willingness to pay of the Sanjiang Plain Wetland and Zhalong Wetland obtained by the random utility model is 124.70 yuan/(person·year)and 143.45 yuan/(person·year),which is higher than the results of by 112.49 yuan/(person·year)and 79.68yuan/(person·year)estimated by the spike model.The evaluation results of the two models are significantly different.The random utility model has low variable significance becanuse a certain proportion of zero-payment samples are excluded.(3)by comparing the values of log-likelihood function and the AIC statistics,the spike model has better fitting effect than the random utility model.Therefore,WTP results estimated by the spike model are selected as the basis of total value accounting,and the ecological protection value of the Sanjiang Plain wetland was 4.312 billion yuan/year in 2014,and 3.056 billion yuan/year in 2018.The research conclusion is helpful to optimize the treatment mode of zero observation value lay a thoretical foundation for improving the effectiveness ans reliability of CVM,and provide a scientific basis for the formulation and implementation ecological environment management policies. |
PDF Full Text Request