Estimation Of Non-market Value Of Prime Farmland Based On Choice Experiment Model

With rapid economic development and acceleration of urbanization process, the loss of farmland in China is very serious. As one of the core issues of farmland protection, the protection of prime farmland bears an important function of protecting farmland. Although central government makes the protection of prime farmland as a basic national policy, and implements the world's strictest protection institutions, the effect of the policy does not reach the expectation, making the protection of prime farmland in trouble. One of the main reasons is that the economic compensation for the protection of prime farmland has not been established. A comprehensive understanding and scientific assessment of the value of prime farmland is an important prerequisite for building the economic compensation for the protection of prime farmland.Prime farmland is an important resource for human survival and development, and its total value includes market value and non-market value. As the non-market value of prime farmland like social value and ecological value cannot be achieved through market transactions, it's difficult to measure it with market price. Therefore, it needs to apply non-market valuation methods to get a reasonable and scientific estimation. So far,evaluation of non-market value of prime farmland in China mainly uses CVM method, although it may occur embedded bias, strategic bias, starting point bias and so on. therefore, the validity and reliability of CVM are still controversial in the academic. Compared with CVM, choice experiment model (CE) can avoid biases caused by CVM, and solve the problem of profit and loss comparison between multi-attributes of prime farmland, that is suitable for the evaluation of non-market value of prime farmland. However, there is no research about introducing CE model to the evaluation of non-market value of farmland and prime farmland in China.First of all, based on the literature review of the study on the theory and methods of non-market value of prime farmland, this paper points out that non-market value is an important part of the total value of prime farmland which cannot be ignored, and proposes the initial attempt of the introduction of CE model to the study of non-market value of prime farmland. And then, this paper takes Deqing County, Zhejiang Province as an example, discusses the cognitive differences of urban and rural residents for the protection of prime farmland based on face-to-face survey, and analyses the cognitive preferences and its possible sources of heterogeneity of urban and rural residents for the non-market value of prime farmland by applying CE model. At last, urban and rural residents'willingness to pay for the non-market value of prime farmland is evaluated and compared, and the non-market value of prime farmland in Deqing County is obtained. At the mean while, this paper investigates the reasons and its influencing factors for respondents'unwillingness to pay.Based on it, this paper concludes that:(1) both urban and rural residents had known the non-market value of prime farmland at some certain, and recognized the need and importance for continuing the protection of prime farmland, however, the urban residents'cognition of non-market value of prime farmland is more comprehensive and profound compared with that of rural residents; (2) urban and rural residents paid more attention to the social value of prime farmland, and the cognitive preferences of urban and rural residents for the non-market value of prime farmland are influenced by age, education, family size, income, behavior and attitude variables; (3) urban residents'willingness to pay is generally higher than rural residents', urban residents'willingness to pay per year is 143.04 Yuan/hu while rural residents'is 27.47 Yuan/hu, and sex, age, family size, income and attitude variables have a major impact on their willingness to pay. Besides, the income of urban and rural residents significantly affects the level of their payment; (4) Mixed Logit model reveals individuals' heterogeneity of preferences and its possible source, and it also shows higher degree of model fit and better explanation than MNL model. However, the estimated results were different in the application of assessing willingness to pay for the property by MNL model and Mixed Logit model; (5) the results show that, in 2009, non-market value of prime farmland in Deqing County is 258,136,900 Yuan, and non-market value per unit is 12670.34 Yuan/ha., accounting for 54.17% of the total value of prime farmland.The policy recommendations of this study are:the government should strengthen the propaganda for non-market value of prime farmland, raise people's awareness for protecting prime farmland and the scarcity of prime farmland; set up special funds for the protection of prime farmland, and establish the mature system of the collection, use and management of funds; establish economic compensation mechanism of the protection of prime farmland with the coordination of related policies, and fully mobilize both residents and local government's initiation for protecting prime farmland.