Willingness to pay for clean air in China

Richard Freeman,Wenquan Liang,Ran Song,Christopher Timmins

doi:10.1016/j.jeem.2019.01.005

Abstract

We use a residential sorting model incorporating migration disutility to recover the implicit value of clean air in China. The model is estimated using China Population Census Data along with PM2.5 satellite data. Our study provides new evidence on the willingness to pay for air quality improvements in developing countries and is the first application of an equilibrium sorting model to the valuation of non-market amenities in China. We employ two instrumental variables based on coal-fired electricity generation and wind direction to address the endogeneity of local air pollution. Results suggest important differences between the residential sorting model and a conventional hedonic model, highlighting the role of moving costs and the discreteness of the choice set. Our sorting results indicate that the economic value of air quality improvement associated with a one-unit decline in PM2.5 concentration is up to $8.83 billion for all Chinese households in 2005.

Highlights

Reliable estimation of household preferences for clean air has long been an important topic for economists and policy makers
We focus on the full model, 25To capture the median household preference for clean air, we use the median PM2.5 concentration for affected households to calculate the marginal willingness to pay (MWTP)
We develop a discrete-choice model of household residential location decisions that incorporates migration disutility and use it to recover the parameters of indirect utilities associated with residing in various cities across China

Summary

Introduction

Reliable estimation of household preferences for clean air has long been an important topic for economists and policy makers. At its heart is a discrete-choice model of household residential location decisions, incorporating migration disutility This empirical model follows Bayer et al (2009), who estimate the economic value of air quality improvement in U.S metro areas; in. No household could improve its utility by migrating to a new city, given income, housing price, local amenities, moving disutility and all the parameters in the equation (6) and (7) (Kuminoff et al, 2013; Klaiber and Kuminoff, 2014). In the case of air quality, local economic activity is likely to be positively correlated with air pollution concentration along with housing price and income This endogeneity problem leads to biased estimates of the marginal utility of air quality. We estimate the residential sorting model and recover counterfactual household income and city-specific housing price using the 2005 One-Percent Population Census of China.

Modeling moving disutility in the first-step estimation

Sensitivity Analysis

Conclusion