Economists have attempted to understand the determinants of healthcare use and expenditure for a long time (Grossman 1972), and a fundamental question has been about the nature of healthcare as an economic good. The expectation that healthcare use would increase disproportionately as income increases (the ‘luxury good hypothesis’) has dominated the debate. Numerous studies have examined this question by quantifying the income elasticity of healthcare (see Costa-Font et al. 2011 for an overview). However, in most studies, the fundamental assumption that income is exogenous is likely to be violated. For instance, individuals in poor health may be less likely to participate actively in the labour market, but at the same time consume more healthcare – the income and healthcare association is unlikely to be causal. Finally, unlike other goods, health is often insured, and healthcare is consumed via insurance products. Hence, extra income might lead individuals to take up health insurance. However, again the mechanisms behind this are largely unobservable, and the effect of income might not be causal.
Unknown effect of income on healthcare
Until recently, one of the most commonly held views in economics has been that healthcare is a luxury good. However, results supporting this view have come from studies examining health expenditure data, rather than healthcare use. This view has additionally been challenged by a study exploiting oil income shocks at an aggregate level (Acemoglu et al. 2013). Further, most of the research to date does not distinguish between preventive and curative health services. Finally, in health systems where there is mainstream public health insurance, a windfall of income might simply lead individuals to switch to private healthcare. It is important to understand how individuals’ decisions about public and private healthcare are determined by income, because these decisions influence support for public sector provision (Propper 2000).
Evidence from lottery wins
In a recent study, we exploit lottery wins as a source of exogenous changes in individuals’ income to obtain causal estimates of lottery income elasticities for healthcare (Cheng et al. 2015). We examine longitudinal data from the UK where about 50% of the population play the lottery, and account for econometric issues such as selection and unobserved heterogeneity. We show that lottery winners with relatively large wins are significantly more likely to choose healthcare from the private sector than from the NHS. The latter is particularly strong for dental care, blood pressure checks, and cervical examinations.
Table 1 Implied income elasticities of health care with respect to lottery winnings and household income
Note: Significance: *** 1%; ** 5%; * 10%. Statistical significance refers to the regression coefficient estimates. Estimates of income elasticity are calculated as percentage change in the proportion of individuals obtaining public or private care versus no care given a 1% increase in lottery winnings or household income.
Our results are important in the context of the luxury good hypothesis, as they suggest that income elasticities for public healthcare services are close to zero – the implied lottery income elasticities for private healthcare suggest that healthcare is a normal good (see Table 1). We find an elasticity of only 0.96 for cervical examinations. Both sets of estimates are similar to those obtained by Kenkel (1994), who, using US data, finds an income elasticity for preventive care of 0.06. Hence, our results are consistent with evidence from microeconomic studies that support the notion that healthcare is a necessity, rather than a luxury good.
Our results indicate that an expansion of income in developed economies is likely to increase the use of private and preventive healthcare but will leave the use of public healthcare largely unchanged. An expansion of healthcare expenditure is likely to come from higher healthcare costs associated with new healthcare technology (Breyer 2011).
Acemoglu, D, A Finkelstein, and M J Notowidigdo (2013) “Income and health spending: Evidence from oil price shocks”, Review of Economics and Statistics, 95 (4): 1079-1095.
Breyer, F, J Costa-i-Font and S Felder (2011) “Does ageing really affect health expenditures? If so, why?”, VoxEU.org, 14 May.
Cheng, T C, J Costa-i-Font and N Powdthavee (2015) “Do you have to win it to fix it? A longitudinal study of lottery winners and their health care demand“, CEPR Discussion Paper, DP1339.
Costa-i-Font, J, M Gemmill and G Rubert (2011) “Biases in the healthcare luxury good hypothesis: A meta-regression analysis”, Journal of the Royal Statistical Society: Series A (Statistics in Society), 174 (1): 95-107.
Grossman, M (1972) “On the concept of health capital and the demand for health”, Journal of Political Economy, 223-255.
Kenkel, D S (1994) “The demand for preventive medical care”, Applied Economics, 26(4): 313-325.
Propper, C (2000) “The demand for private healthcare in the UK”, Journal of Health Economics, 19(6): 855-876.
This article is published in collaboration with Vox EU. Publication does not imply endorsement of views by the World Economic Forum.
To keep up with the Agenda subscribe to our weekly newsletter.
Author: Terence Cheng is a Senior Lecturer, School of Economics, University of Adelaide. Joan Costa-i-Font is an Associate Professor (Reader) of Political Economy, London School of Economics and Political Science. Nattavudh Powdthavee is a Professorial Research Fellow, Melbourne Institute of Applied Economic and Social Research; Principal Research Fellow, Centre for Economic Performance, LSE.
Image: A nurse poses for a photo in a trauma center of the University of Mississippi Medical Center in Jackson. REUTERS/Jonathan Bachman