How Do Chinese Smokers Respond to Cigarette Price? Evidence From a Nationwide Household Survey

Despite the Chinese government's implementation of tax increases in 2009 and 2015 to curb tobacco use, the anticipated decline in cigarette sales has not materialized. This suggests a need for a deeper understanding of smokers' demand behavior. The present study aims to provide a more accurate estimation of the price and income elasticities of cigarettes in China.

We utilized Deaton's model to analyze a comprehensive micro-level dataset comprising 10 892 households surveyed in the 2012 wave of the China Family Panel Survey, with the aim of estimating the price and income elasticities of cigarette demand, as well as to investigate how these elasticities vary across different income brackets. These resulting elasticities were subsequently integrated into the World Health Organization Tax Simulation Model to simulate the potential impacts of various tax increase scenarios on cigarette consumption and tax revenue.

The price elasticity in China hovers around -0.72, with an income elasticity of about 0.23. Heterogeneity analysis indicated that lower-income smokers exhibit greater sensitivity to price changes. Policy simulations suggested that tax increases would substantially reduce consumption while concurrently increasing tax revenue.

Chinese smokers exhibited a considerable sensitivity to cigarette prices, underscoring the efficacy of taxation as a policy instrument for shaping smoking behavior. As tax reforms continue and cigarette retail prices increase, there lies promising potential for achieving tobacco control goals, generating additional revenue, and improving social equality.

A primary contribution of this study is the provision of a more reliable estimation of cigarette price and income elasticities in China. These parameters are crucial for assessing the potential impact of tobacco tax increases on cigarette consumption and government revenue. By leveraging a nationally representative survey from 2012 and employing Deaton's identification strategy, we significantly enhanced the data quality and methodological rigor of our analysis, leading to more accurate estimation results compared to previous studies.