Adverse selection in China's New Rural Cooperative Medical Scheme

Lijian Qin, Suwen Pan, Chenggang Wang, Zhongyi Jiang

Research output: Contribution to journalReview articlepeer-review

4 Scopus citations


Purpose - The purpose of this paper is to examine the adverse selection in participation in the New Rural Cooperative Medical Scheme (NRCMS), as well as in outpatient and inpatient service utilization, in Chaoyang, Beijing, China. Design/methodology/approach - Probit model is established to test whether the rural Hukou family member in Combined Household (CH) is statistically different from the Pure Rural Household (PRH) in enrollment in NRCMS. Seemingly Unrelated Regression (SUR) model is adopted to examine the difference in the utilization of outpatient and inpatient between the rural Hukou family members in the two kinds of households. Findings - This paper finds that the rural Hukou family member in CH has more probability to enroll in NRCMS than the counterpart in PRH. In the period of six months, the rural Hukou family member in CH exceeds PRH by 0.73 times in outpatient visit number per capita. The former average spends yuan 157 more in outpatient service and is reimbursed yuan 53 more from NRCMS than the latter. Moreover, on average, rural Hukou family member has no difference in the inpatient service utilization between the two kinds of households in the period of 12 months. Originality/value - This is the first study to empirically test the adverse selection in China's medical insurance market from the perspective of two different types of households, which are CH and PRH.

Original languageEnglish
Pages (from-to)69-83
Number of pages15
JournalChina Agricultural Economic Review
Issue number1
StatePublished - 2012


  • Adverse selection
  • China
  • Health insurance
  • Health services
  • Medical insurance
  • Outpatients
  • Patients
  • Rural regions
  • Suburban district


Dive into the research topics of 'Adverse selection in China's New Rural Cooperative Medical Scheme'. Together they form a unique fingerprint.

Cite this