Can government subsidies and public mechanisms alleviate the physical and mental health vulnerability of China's urban and rural residents?

Abstract

Background: Poverty vulnerability has been defined as the likelihood of a family falling into poverty in the upcoming months. Inequality is a major cause of poverty vulnerability in developing countries. There is evidence that establishing effective government subsidies and public service mechanisms significantly reduces health poverty vulnerability. One of the ways to study poverty vulnerability is by using empirical data such as income elasticity of demand to perform the analysis. Income elasticity refers to the extent to which changes in consumers' income affect changes in demand for commodities or public goods. In this work, we assess health poverty vulnerability in rural and urban China. We provide two levels of evidence on the marginal effects of the design and implementation of government subsidies and public mechanisms in reducing health poverty vulnerability, before and after incorporating the income elasticity of demand for health.

Methods: Multidimensional physical and mental health poverty indexes, according to the Oxford Poverty & Human Development Initiative and the Andersen model, were implemented to measure health poverty vulnerability by using the 2018 China Family Panel Survey database (CFPS) as the data source for empirical analysis. The income elasticity of demand for health care was used as the key mediating variable of impact. Our assessment was conducted by a two-level multidimensional logistic regression using STATA16 software.

Results: The first level regression indicates that the marginal utility of public mechanism (PM) in reducing urban and rural vulnerability as expected poverty on physical and mental health (VEP-PH&MH) was insignificant. On the other hand, government subsidies (GS) policies had a positive suppression effect on VEP-PH&MH to a relatively low degree. The second level regression found that given the diversity of health needs across individual households, i.e., the income elasticity of demand (HE) for health care products, PM and GS policies have a significant effect in reducing VEP-PH&MH in rural and urban areas. Our analysis has verified the significant positive impact of enacting accurate GS and PM policies on effectively reducing VEP-PH&MH in rural as well as urban areas.

Conclusions: This study shows that implementing government subsidies and public mechanisms has a positive marginal effect on reducing VEP-PH&MH. Meanwhile, there are individual variations in health demands, urban-rural disparities, and regional disparities in the effects of GS and PM on inhibiting VEP-PH&MH. Therefore, special consideration needs to be given to the differences in the degree of health needs of individual residents among urban and rural areas and regions with varying economic development. Furthermore, considerations of this approach in the current worldwide scenario are analyzed.

Keywords: Andersen model; Government subsidies; Multivariate logistic regression analysis; Physical and mental health poverty vulnerability index; Public mechanisms.

Fig. 1

Sample size determination (n = 8831) from the 2018 China Family Panel Studies (CFPS) survey

Fig. 2

Difference between urban and rural EII and MII in the eastern, central, and western regions of China in 2018 (EII = Economic Inequality Index, MII = Medical Inequality Index). Note: Based on the methods presented by [65, 66], the equivalence is shown by: $g_{\mathit{ME}}\left(X,,,W\right)=-E\left[log,p,\left(g,(,W,x,)\right)\right]\triangleq H(z)=E\left[{\sum }_{i=1}^m,log,p,\left(y_i\right)\right]+log\left|det,W\right|-E\left[log,p,(,x,)\right]\propto Tlog\left|det,W\right|+{\sum }_{t=1}^T{\sum }_{i=1}^{m}logp\left(w_i,x_t\right).$ Where the maximum entropy (ME) based ICA method meets the perspective of a neural network by estimating the demixing matrix W_ME, which maximizes the entropy H(·) of the nonlinear outputs z = g(y) of a neural network. In particular, using the cumulative distribution function for a nonlinear function g(·), i.e., $g_i^{\prime }\left(y_i\right)=p\left(y_i\right)$, assures the equivalence between the ME method using W_ME

Fig. 3

The gap of the urban-rural economic index and public health service index for each province and city in the three regions of China. Note: Based on the availability of data, only thirty provinces of China are shown in the figures for this study. Based on the methods presented by [65, 66], the equivalence is shown by: $g_{\mathit{ME}}\left(X,,,W\right)=-E\left[log,p,\left(g,(,W,x,)\right)\right]\triangleq H(z)=E\left[{\sum }_{i=1}^m,log,p,\left(y_i\right)\right]+log\left|det,W\right|-E\left[log,p,(,x,)\right]\propto Tlog\left|det,W\right|+{\sum }_{t=1}^T{\sum }_{i=1}^{m}logp\left(w_i,x_t\right).$ Where the maximum entropy (ME) based ICA method meets the perspective of a neural network by estimating the demixing matrix W_ME, which maximizes the entropy H(·) of the nonlinear outputs z = g(y) of a neural network. In particular, using the cumulative distribution function for a nonlinear function g(·), i.e., $g_i^{\prime }\left(y_i\right)=p\left(y_i\right)$, assures the equivalence between the ME method using W_ME