Does energy poverty increase health care expenditures in China?

ABSTRACT

Using the 2012–2018 waves of the China Family Panel Studies, we investigate the impact of energy poverty (EP) on health care expenditures among Chinese adults aged 18+. Employing a methodology combining a random effects two-part model and instrumental variable estimations, we show that EP leads to higher levels of total (305 yuan/year), out-of-pocket (199 yuan/year), inpatient (230 yuan/year) and other (113 yuan/year) health care expenditures, with more pronounced impacts among females and those living in urban areas and Eastern China. These results are robust not only to alternative EP and health care expenditure measures but also to a series of estimation approaches that control for endogeneity. An additional structural equation modelling analysis of the underlying pathways further reveals that this EP-health care expenditure relationship is mediated by individual self-reported health as well as expenditures on food and other daily necessities. Combating EP is an effective way to improve people’s health and reduce the burden on health care expenditures. Policymakers should also pay more attention to vulnerable groups such as women.

KEYWORDS:

• Energy poverty
• health care expenditures
• Random effects two-part model; China

JEL CLASSIFICATION:

• I10
• I11
• I32
• Q40

Acknowledgments

We are very thankful to the Institute of Social Science Survey at Peking University for providing the CFPS data used in this study. This work was supported by the National Natural Science Foundation of China (grant numbers 71804142 and 72074178). We thank the editor Mark Taylor and two anonymous referees for their valuable comments on an earlier version of this paper.

Data availability statement

The data that support the findings of this study are openly available in [Peking University Open Research Data] at https://opendata.pku.edu.cn/dataverse/CFPS;jsessionid=1e0dcab23e9373f51acc4489578f.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ This initiative gives priority to health and aims to improve national health policy and ensure the delivery of comprehensive lifecycle health services for people.

² A detailed discussion of the differences and similarities between EP and FP is available in Li et al. (2014).

³ Our study is different from Kahouli (2020) because we focus on the relationship between EP and health care expenditures rather than health. In our case, we take health as one possible channel for the linkage between EP and health care expenditures.

⁴ In this study, South Asia covers six countries, including Afghanistan, Pakistan, India, Bangladesh, Nepal and the Maldives..

⁵ Information on energy utilization in 2010 is unavailable..

⁶ In our dataset, of the four types of health care expenditures, zero health care expenditures account for nearly 45% of our sample on average, and the distribution is highly skewed..

⁷ Since fixed effects (FE) estimates are biased in nonlinear models with small group sizes (with a few exceptions where conditional maximum likelihood estimators exist) (Jiang and Ni 2020), we implement RE estimation rather than FE estimation..

⁸ Energy expenditures are unlikely to account for a large share of household budgets and are thus unlikely to significantly affect other expenditures (Churchill and Smyth 2021). In China, the average share of energy expenditures in household income is approximately 7%-8% (Cheng, Tani, and Wang 2021)..

⁹ Liquid fuels include domestic heating and lighting oil; such fuels are carbon-intensive energy alternatives similar to solid fuels (Lima, Ferreira, and Leal 2021).

¹⁰ Household health care expenditures are household total direct health care expenditures (excluding that was reimbursed or reimbursable but including that was paid by or borrowed from relatives) in the previous year. We control for the age and education of the household head in the regression.

¹¹ The information on the population aged 18 and over is available in http://www.ncjggw.gov.cn/index.php?m=content&c=index&a=show&catid=37&id=7475.

Additional information

Funding

The work was supported by the National Natural Science Foundation of China [71804142,72074178].