意愿调查法中的偏好不确定性研究综述

对非市场物品和服务,如环境改善和资源保护,进行经济价值评估是学术界面临的一大难题。意愿调查法（Contingent Valuation Method, CVM）作为非市场物品经济价值评估的一种重要方法,其方法论在不断地接受检验并得到改进,近年来一个重要进展是其分析框架由确定偏好假设向不确定偏好假设转变。为系统认识国际上对CVM中受访者偏好不确定性的研究进展、推动国内相关研究,本文在简单介绍产生不确定结果的原因和不确定偏好假设下标准CVM不足的基础上,系统梳理了不确定偏好的测度方法和不确定偏好下价值计算方法,分析了各种方法的优劣以及考虑不确定性对价值估计值大小和估计效率的影响。总体来看,不确定偏好假设已基本形成共识,个人支付意愿应为随机变量也得到广泛支持。多数学者认为借助一定的测度技术获取的不确定性信息是有价值的,而如何更准确地测度和利用偏好不确定程度以提高资源环境等非市场物品价值估计效率应是今后研究的重点。最后,针对既有研究存在的问题,从加强个人偏好及不确定性的异质性研究、改进问卷设计等方面提出对未来研究工作的建议。

A review of preference uncertainty in contingent valuation method

Economic valuation of non-market goods and services, such as improvement of environmental quality and conservation of natural resources, is a methodologically strong challenge to economists. As a primary method of economic valuation of non-market goods, the contingent valuation method, CVM, has been continuously tested and improved. One of the improvements is the change of assumption from certain preferences to uncertain preferences. This article first summarizes the reasons of uncertain preferences and potential problems caused by ignoring the uncertainty nature of preferences in contingent valuation studies. In order to have a better understanding about the issues, this article also provides a systematic review of the measurement techniques of uncertain preferences and willingness-to-pay (WTP) estimation models where the respondents’ uncertain information is incorporated. Both advantages and disadvantages of different methods are summarized. Also analyzed are the impacts of incorporating uncertain preferences on the magnitude of the estimates as well as the estimation efficiency. In general, there is a consensus among researchers on the uncertainty characteristic of respondents’ preferences. The opinion that individuals’ WTP should be a random variable with a specific distribution is generally accepted. Most of the researchers believe that information about uncertainty obtained through specific techniques is valuable for calibrating WTP estimates and more attention should be paid to the accuracy of uncertainty estimation to improve the efficiency of WTP estimation. Suggestions about future research, such as studying on heterogeneity of individuals’ uncertain preferences and improving the survey design and implementation, are provided at the end of the article.