Skip navigation to content
eriu: Economic Research Initiative on the Uninsured Initiating and disemminating research to spark new policy discussion on health coverage issues.
Fast Facts  
Funded Research Home

      Sort by Author (A-Z)
      Sort by Topic


      Costs of Health Insurance
      Demand for Health Insurance   
      Employment-Based Coverage
      Labor Markets
      Near Elderly Population
      Vulnerable Populations
      Welfare, Medicaid, and SCHIP

Find a Document

      Research Highlights
      Research Findings
      Working Papers
      Q & A with the Author
      Conversations with       Economists

Home > Funded Research Home > All > Sort by Author (A-Z) > Hill & Kreider

ERIU Funded Research Projects

Sort by: Author (A-Z) | Topic

A | B | C | D | E | F | G | H | I | J | K | L | M | N | O | P | Q | R | S | T | U | V | W | X | Y | Z

Author: Hill, Steven C. ; Kreider, Brent
Working Paper: Partially Identifying Treatment Effects with an Application to Covering the Uninsured (PDF) ; April 2006

We extend the nonparametric literature on partially identified probability distributions and treatment e¤ects. We then use our analytical results to bound the impact of universal health insurance on provider visits and medical expenditures. Our approach accounts for uncertainty about the reliability of self-reported insurance status as well as uncertainty created by unknown counterfactuals. As part of the contribution, we provide sharp bounds on the conditional mean of a random variable for the case that a binary conditioning variable is subject to arbitrary endogenous measurement error.

Using data from the 1996 Medical Expenditure Panel Survey (MEPS), we construct health insurance validation data for a nonrandom portion of the sample based on insurance cards, policy booklets, and follow-back interviews with employers and insurance companies. Under relatively weak verification and monotonicity assumptions, we estimate that monthly per capita provider visits under universal coverage would rise by no more than four-tenths of a visit (a 9% change) across the nonelderly population and that per capita expenditures would rise by less than 15%.