Data measuring the options available to consumers or their
behavior over time—and with sufficient observations results—are rare. As a result, available elasticity variable bias or endogeneity bias.
to obtain statistically significant estimates often reflect omitted
Many statistical models have been used to address
methodological challenges due to limited data. However, the from improvements in techniques for linking multiple sources consumer response in complex markets and over time.
field would benefit of data to estimate
Limited knowledge about the demand for certain insurance products, such as high- deductible health plans and health savings or reimbursement accounts. While high-deductible insurance products have captured the attention of policymakers as a promising model for reducing cost and expanding coverage, there are few analyses of the potential demand for these products by the general public or the change in their use of care once enrolled.
Limited knowledge about demand among critical subpopulations, such as Medicare beneficiaries, low-income populations, and individuals or families with income just above the level that typically would qualify them for public assistance. Demand elasticities estimated for the general population are likely to produce flawed estimates of the impacts of policy changes that target particular subpopulations.
Limited knowledge about the elasticity of demand for specific services—especially mental health care and long-term care. These services represent a substantial share of the cost of state Medicaid programs, in particular—while private insurance historically has restricted coverage for mental health care and long-term care relative to other service types. Consumer responses to improved coverage for these and other types of care—such as preventive services or specific types of prescription drugs— merit further research to support improvements in the design of public and private coverage.
Absence of protocols for applying elasticity estimates in policy simulations. Inappropriate use of estimated elasticities (including indiscriminate use without reference to the formulas that calculated them) can introduce substantial error in the analysis of policy options. In addition, failure to conduct sensitivity analyses can suggest more certainty about effects than is warranted by the variance around estimates of demand elasticity. A clearer and common understanding of how specific elasticity estimates are formulated and how analysts should handle error in the estimates could help to reduce the range of disagreement around public policy options.
Bridging any or all of these gaps would contribute significantly to improving the science of elasticity estimation and the resources available to understand the potential impacts of major changes in health care policy.