used the 1989 National Long Term Care Survey to examine demand for nursing home services
among the elderly population. He found that income elasticity of demand for nursing home
services among private-pay patients and patients eligible for Medicaid was not significantly
different from zero.
At least two recent studies have attempted to estimate income elasticity by observing
differences in average income and health care spending across states or nations, and over a
longer time period (Dimatteo 2003, Freeman 2003). Both found that the income elasticity of
demand for health care was greater at higher levels of aggregation (e.g., national versus
individual expenditures), and also that the income elasticity of demand among consumers at low
income levels is greater than at high income levels.
The consensus of the limited literature estimating the income elasticity of demand for either
health insurance or health services is that both are income-inelastic. However, because few
studies actually observe an exogenous change in consumers’ levels of income, nearly all studies
that estimate income elasticity are based on observational data. These studies suggest that as
income increases, the demand for health insurance and health services also increases, but the
magnitude of the increase is relatively small: the estimates of income elasticity are consistently
in the range of zero to 0.2 with respect to health care services, and zero to 0.1 with respect to
aggregated state- or country-level data have produced higher estimates of income elasticity—but
all in the range of 0.2 to 1.5.
(continued) and illustrates the importance of recognizing unobserved alternatives when considering estimates of income elasticity.