Parental eligibility for public health insurance: a study of the State Children's Health Insurance Program and child coverage rates

Abstract

This article investigates whether offering parental coverage for public health insurance increases coverage rates among children. Using data from the Current Population Survey (CPS), we examine the effect of parental benefits on participation rates of children in the State Children's Health Insurance Program (SCHIP). Our findings indicate a significant difference in take-up rates for children in states offering parental coverage compared to neighbouring states that did not in the period surrounding the introduction of the SCHIP program (1997–2002). Specifically, states offering parental coverage have greater program participation (on the order of 10% points) among children even after controlling for observable demographic characteristics and the household's position in the income distribution. Thus, extending coverage to the parents of SCHIP-eligible children appears to be an effective strategy to reduce uninsurance rates among children.

Notes

¹ Medicaid or SCHIP.

² A study by Dubay and Kenney (2003) examining extensions of parental eligibility for Medicaid found that the expansions had a positive effect on child coverage rates in the state of Massachusetts.

³  PublicCoverage equals one in 1998 if the child is covered by Medicaid, or more specifically, if the CPS variable ‘chmc’ equals one. There should not be children included in the sample who have public health insurance through SCHIP in 1998 because the expansions had yet to be implemented. However, there is a small amount of measurement error. PublicCoverage equals one in 2003 if ‘chmc’ equals one or if the child is covered by SCHIP in which case the CPS variable ‘pchip’ equals one.

⁴ Specifically, SCHIP equals one for all observations from the CPS in 2003 when SCHIP was available. SCHIP equals zero for all observations from the CPS in 1998 prior to the programs inception.

⁵ The treatment groups are New Jersey and Wisconsin as they extended SCHIP benefits to parents while the control groups are Pennsylvania and Illinois.

⁶ The family-level poverty variable  = income as a % of the Federal Poverty Level, where ‘familyinc’ is a CPS variable indicating a family's income and ‘povcut’ is another CPS variable that provides the specific poverty level cutoff for a given family based on the number of family members.

⁷ Having public health insurance coverage is defined by our dependent variable ‘PublicCoverage’.

⁸ The bad health variable is a recoded dummy variable from the CPS variable ‘health’ that is equal to one if the individual indicated they had poor health.

Table 2. SCHIP eligible children: mean values for the dependent and explanatory variables, by state

⁹ To ensure clarity on this issue, the sample of SCHIP eligible children for 1997 and 2002 include all children who were eligible for SCHIP in 2002 and all children in 1997 who would have been eligible for SCHIP based on the same 2002 criteria.