Consideration of Supply-Side Behaviors
Little is known about how insurers or providers might respond to a large change in the price
for coverage or services, such as might occur if a sizeable refundable tax credit for health
insurance were enacted. In such an event, it is reasonable to expect that insurers might change
the nature of insurance products in an attempt to offer “zero premium” plans (net of the
refundable credit) and that providers might alter the cost of seeking care (to attract newly insured
patients or avoid bad debt that may be associated with underinsured patients). Either response
would alter demand.
While including such supply-side behaviors in a structural model can raise methodological
challenges, understanding supply-side behaviors is essential to anticipating the long-term effects
of policy change. At minimum, researchers should consider possible scenarios of supply-side
changes and conduct sensitivity analyses of demand changes in response to these scenarios.
However, at present, there is virtually no basis for selecting among or quantifying possible
scenarios. Even a modest program of research on supply-side responses could be very helpful in
guiding how researchers might consider supply-side effects when modeling policy change.
B. SPECIFIC INSURANCE PRODUCTS, SERVICES, AND SUBPOPULATIONS
Both supply and demand in the health care market are heterogeneous: not only are products
tailored to address different needs, but the population varies widely with respect to attributes that
affect decision-making about health insurance coverage and service use. To be efficient, public
policy initiatives to expand coverage probably will be targeted narrowly to specific
subpopulations. Consequently, for public policy to be effective, it is necessary to anticipate how
target subpopulations will respond. Our assessment of the literature suggests there are a number
of products, services, and subpopulations that may be important targets for public policy but
have not been studied at all or studied adequately.