What Determines Health Care Expenditure in High Income Countries?

OHE’s Yan Feng and Grace Hampson, along with colleagues from the Health Foundation and RAND Europe, have published a new paper in Applied Economics and Finance which explores the factors that determine health care expenditure in OECD countries.

A new journal article entitled “What Determines the Health Care Expenditure of High Income Countries?”  has been published online in Applied Economics and Finance.

This article is authored by OHE’s Dr Yan Feng and Grace Hampson, in collaboration with Anita Charlesworth, Adam Roberts and Toby Watt from the Health Foundation, and Jon Sussex from RAND Europe.

Controlling the rise in health care costs continues to be a major focus of health care policy. It is important for governments to understand what is driving the rise in health care expenditure and what the impact will be over the coming years. This paper utilises an econometric model to understand the determinants of health care expenditure.

Data from the OECD and IMS databases for 18 OECD countries between 1988 and 2012 are analysed at the year and country level. The study applied three different methods: (1) panel data models with country fixed effects; (2) a first difference model; and (3) a vector error correction model to account for the long run and short run effects as well as the endogeneity of the explanatory variables.

The empirical results suggest that the use of different econometric specifications has a significant impact on both establishing the determinants of health expenditure and their magnitudes.

Based on results from the “best” model in the paper (the vector error correction model), we find that gross domestic product (GDP) is the key driver for health care expenditure growth. A 1% increase in GDP is associated with a 1.1% increase in the health care expenditure.

This study provides empirical evidence about the determinants of health expenditure in high income OECD countries, as well as their magnitudes, based on recent data. The results suggest a need for caution when selecting modelling specifications for exploring health care expenditure patterns.

Characteristics of the data that should be considered when selecting the model specification include the dynamic nature of time series data as well as the causal relationship between variables. Given the importance of health care expenditure in decision making, more research in this area should explore these issues further.

This project was funded by a research grant from the Health Foundation. Views expressed in the paper are those of the authors and are not necessarily those of the Health Foundation.

The full article is available here [open access]

For more details please contact Dr Yan Feng.