Joshua Witter and Dr. Joseph Price, Economics Department
Introduction
I was awarded an ORCA grant to research the unique health insurance market that the children of teen mothers face. I became concerned that these children may not have equal access to private insurance coverage and as a result were at risk of being uninsured or over-enrolling in public insurance programs such as Medicaid. The majority of these children are living in three-generation households where their grandparents are the primary financial provider for the household members. In many instances, but not all, insurance companies will not permit dependent grandchildren to be added to the grandparent’s health insurance plan, especially if the mother still holds primary custody. Switching custody is a significant ordeal and would likely only occur if the mother was not part of the home anyway. To analyze this issue specifically I narrowed my focus to children who are living in homes with their dependent teen mother and her parents, who are providing support for the teen mother and child.
How Many Children of Teen Mothers are there and what kind of insurance do they have?
The American Community Survey (ACS), a nationally representative data set of households sponsored by the US Census, provided the perfect framework to study these teen mothers. Two other datasets I tried working with had too few teen mothers in the sample which made analysis difficult. I used the data years 2008 through 2012 to accumulate a sample of 5813 Children of Teen Mothers (COTM) age 0-6. Breaking this sample number down by year and then using population weights, that number translates to about 130,000-150,000 COTMs age 0-6 in living in the United States in any given year since 2008.
Narrowing down the data to just children who receive primary support from their grandparents, I calculated that 64% of the COTMs were receiving this kind of support. This translates to about 85,000-100,000 children in a given year. It was necessary to narrow down to this group to understand what types of insurance they were enrolling in. The insurance enrollment rates were as follows: 79.3 % were enrolled in Medicaid, 10.5% were enrolled in employer provided health insurance, 1.5% were enrolled in privately purchased insurance, 7.6% were uninsured, and 1.1% were enrolled in some other type of insurance. It is easy to see that these children are overwhelmingly enrolled in Medicaid. The landscape is significantly different for the mothers however. Teen mother enrollment is as follows: 50.1% were enrolled in Medicaid, 29.4% were enrolled in employer provided health insurance, 3.2% had privately purchased insurance, 15.0% were uninsured, and 2.5% had some other type of insurance.
It is easy see that there is a significant disparity between what types of insurance the child has versus that of the mother. In fact I calculated further that if a child’s mother is enrolled under employer sponsored insurance, they have Medicaid 58.6% of the time, and employer insurance only 30.8% of the time. Some of this is attributed to different poverty eligibility limits for different age groups, but nonetheless gives significant proof the system currently promotes a “funneling effect” into Medicaid regardless of the insurance that grandparents have to provide.
Government Expenditures
This enrollment would not be of concern if in general these households fall under the poverty limits established by federal and state governments. So in order to measure the number of COTMs that receive Medicaid with household incomes over the limit, I use the poverty level cutoffs for all the states in 2014 (the most generous cutoffs).1 The levels are different for certain age groups and adults in each state. These differences were accounted for in my analysis.
After implementing the cutoffs for each state and then retrospectively looking to see if they would have qualified in the years reported, I found that 64.3% of COTMs had Medicaid, despite having grandparent support and an ineligible household income. This translates to about 23,000-30,000 COTMs in any given year.
So the question becomes what if these children were able to enroll under private insurance plans? There would undoubtedly be a load of savings in terms of government expenditures. I calculate a general estimate of these savings using Medicaid spending per child (divided by state) as reported by the Kaiser Family Foundation.2 Annual Medicaid spending per child across all states was $2502 (in 2011 dollars), but ranged from $1594 in Wisconsin to $5193 in Vermont. Utah spending per child was slightly below average at $2406. I assigned these costs to the children in the sample according to their respective states and found that expenditures on average would have ranged from $52 million to $74 million in a given year or $325 million total across 2008-2012.
Factors that contribute to having Medicaid
For the last part of my analysis I used logistic regression analysis to determine the factors that contribute the most to children having Medicaid. In a nutshell regression analysis teases out the effect of a characteristic or quality like gender on whether COTMs have Medicaid.
After controlling for poverty levels, I found the following results. White COTMs and COTMs of mixed race were the most likely to be enrolled in Medicaid. Blacks were 15% less likely than whites, and mixed races were at least 48% more likely than whites. Additionally female children were 11% more likely to have Medicaid. Also children who had their care provided by grandparents were almost equally likely to have Medicaid versus those who had care provided by another household member.
Conclusion
Through this project I was able to confirm that there is a significant health insurance market inefficiency for children of teen mothers. Certain demographics are more likely to have Medicaid than others, even after controlling for household income level. Additionally, many children who could have health insurance through a grandparent-sponsored plan are often defaulted into Medicaid, given that purchasing private coverage is more costly. Unfortunately, the government picks up the tab. My suggestion is that health insurance companies should be mandated to cover individuals who can be claimed as a dependent on the grandparent’s tax form. Insurance companies would request documents proving dependency. When the teen mother can claim to be an independent for the first time she can seek out a plan for her child that is contingent on her personal income. I feel that this solution would minimize additional paperwork and allow for more effective expenditures on COTMs and their families. The government could redistribute the savings to help teen mothers with more helpful services like child care during school or education programs, therefore protecting the supporting grandparent’s income and giving teen mothers a way to finish their high school degrees.
1 These eligibility levels can be found at http://www.medicaid.gov/AffordableCareAct/Medicaid-Moving-Forward-2014/Downloads/Medicaid-and-CHIP-eligibility-Levels-Table.pdf
2 Medicaid spending per enrollee is reported in 2011 dollars. The data can be found at http://kff.org/medicaid/state-indicator/medicaid-spending-per-enrollee/#notes