Demand for health: An empirical model of health production in China

As non-communicable diseases and obesity-related health risks become major public health concerns globally, there is an increased need to quantify the determinants of adverse health conditions and the contextual factors that influence these determinants. This paper estimates the production of health and the demand for health inputs using data from the China Health and Nutrition Survey. The empirical model consists of jointly estimated reduced form demand equations for health inputs (i.e., smoking, exercise, alcohol consumption, medical care, and calorie consumption) along with structural health production equations for health outcomes (i.e., self-rated health and body fat) that include these health inputs. The estimation method controls for the endogeneity of health inputs in the health production functions using a semi-parametric maximum likelihood estimator. Estimates from the health production function demonstrate a beneficial impact of exercise on subjective health for both men and women, while also lowering body fat among women. Smoking and drinking are found to be detrimental to subjective, self-rated health for women. Estimates of the determinants of health input demand identify pathways through which changes in the prices of goods, or services like medical care, can modify health behaviors and, subsequently, health outcomes.