Do Children of Immigrant Parents Assimilate into Public Health Insurance? A Dynamic Analysis (Text Version)

On September 15, 2009, Julie Hudson and Yuriy Pylypchuk made this presentation at the 2009 Annual Conference. Select to access the PowerPoint® presentation (741 KB).

---

Slide 1

Do Children of Immigrant Parents Assimilate into Public Health Insurance? A Dynamic Analysis

By
Julie Hudson
Yuriy Pylypchuk

Slide 2

Background

• Insurance status among children with native parents (2005)
  • 67% private.
  • 28% public.
  • 4% uninsured.
• Insurance status among children with at least one foreign born parent (2005)
  • 48% private.
  • 40% public.
  • 12% uninsured.

Slide 3

Background

• Children of immigrant parents are more likely to participate in public coverage than natives by 12 percentage points
  • Consistent with immigrant participation in other public programs (AFDC, Food Stamps).
• Public issue
  • Cost implications for Medicaid and SCHIP.
  • Lack of coverage prevents access to care among children => future burden?

Slide 4

Key Issue

• Does a child's participation in public coverage depend on his/her parent's stay in the U.S?
  • The propensity to participate can increase, decrease, or remain the same with parents length of stay in the U.S.
  • If decreases => children of immigrant parents assimilate out of public coverage.
  • If increases => children of immigrant parents exhibit increasing welfare dependence (found among adults for AFDC/TANF).

Slide 5

Study Objectives

• Does a parent's length of stay in the U.S affect eligible children's propensities to
  • Enter into public coverage.
  • Retain public coverage.
• Are there differential effects of immigration- citizenship status of Mothers versus Fathers?
• What is the role of a child's own immigration- citizenship status for entry/retention?
• How do immigration characteristics affect a child's participation in public coverage in the steady state.

Slide 6

Literature

• Borjas and Trejo (1991), Wei-Yin Hu (1997)
  • Adult immigrants assimilate into welfare programs in the U.S.
• Hanson and Lofstrom., 2003
  • Immigrants assimilate out of welfare in Sweden.
• Currie 2000, Buchmuler et al., 2008
  • SCHIP expansion increased participation in public coverage among children with foreign born household heads.
• Ham et al., 2008.
  • Transitions among private, pubic, and no insurance. No immigrants characteristics in the model. Hispanics are more likely to enter public coverage.

Slide 7

Data

• Medical Expenditures Panel Survey (MEPS), Years 1996-2005, panels 1-9.
• 0-17 year old children.
• Eligible for Medicaid or CHIP.
• Model quarterly transitions from and to public coverage over two year period
  • Exclude
    • Children with missing coverage for 3 months in a row.
    • Children who appeared in survey for 3 months or less.
    • Children without any parent or head of the household.
  • Eligibility criteria varies annually.

Slide 8

Data

• Information about immigrants is extracted from National Health Interview Survey.
• In all models we control for
  • State quarterly unemployment rate.
  • Cohort effects.
  • Parent education and health status.
  • Children's health status, region and MSA, race and age.
• All models are estimated separately for two and one parent households.

Slide 9

Model

• Hazard of entering public coverage.
• M - years of stay in the U.S.
• Hazard of retaining public coverage
  • Do not control for initial conditions.
  • Do not control for unobserved heterogeneity.

Slide 10

Steady state

• Let R be 2X2 transition matrix, where the element of the matrix, represents the predicted probability of moving from state k to state j.
• Let P denote the row vector of participating in public coverage or having other insurance status in steady state.
• To find element in matrix P, solve.
• The effect of the binary covariate, X, on steady-state probability is simply.
• The expression informs us of the long run effects of a specific covariate on the likelihood of being in public coverage.

Slide 11

Parent Characteristics of Eligible Children by Parental Nativity & Stay in the U.S.

	Both Native	All Immig	Immig <5 yrs	Immig 5-10 yr	Immig 10-15 yr	Immig >15 yr
Mother
Offer	0.43	0.29**	0.16**	0.20**	0.25**	0.39*
HS deg	0.68	0.36**	0.36**	0.28**	0.34**	0.44**
Empl	0.53	0.39**	0.34**	0.33**	0.37**	0.46**
Father
Offer	0.71	0.52**	0.43**	0.43**	0.49**	0.59**
HS deg	0.65	0.36**	0.41**	0.30**	0.33**	0.44**
Empl	0.83	0.87**	0.81	0.88**	0.90**	0.86

Significantly different from Both Native: * 10% **5% ***1%

Slide 12

Transition Matrix Mean Probability of Transition

	Both Native	Immig <5 yrs	Immig 5-10 yr	Immig 10-15 yr	Immig >15 yr
	Mother is an Immigrant
Entry into Public Coverage	0.03	0.05**	0.06**	0.07**	0.05**
Retention of Public Coverage	0.95	0.96**	0.95	0.95	0.95
	Father is an Immigrant
Entry into Public Coverage	0.03	0.06**	0.07**	0.06**	0.05**
Retention of Public Coverage	0.95	0.96	0.95	0.95	0.95

Slide 13

Discrete Hazard Results: Father's Stay - 2 Parent HH

	Citizen Immigrant Father		Non-Citizen Immigrant Father
	Entry	Retention	Entry	Retention
<5 years	-0.013	-0.015	0.002	-0.004
5-10 years	-0.012	-0.012	0.002	-0.002
10-15 years	-0.023**	-0.010	-0.013	-0.001
>15 years	-0.019*	-0.014	-0.008	-0.004

& Comparison group: Children with two Native Parents.

Slide 14

Discrete Hazard Results: Mother's Stay - 2 Parent HH

	Citizen Immigrant Father		Non-Citizen Immigrant Father
	Entry	Retention	Entry	Retention
<5 years	0.011	0.012	0.009	0.002
5-10 years	0.019	-0.000	0.018	-0.013
10-15 years	0.029	-0.014	0.027*	-0.030**
>15 years	0.019	-0.028*	0.017	-0.048**

& Comparison group: Children with two Native Parents.

Slide 15

Discrete Hazard Results: Other Characteristics - 2 Parent HH

	Entry	Retention
Child
Immigrant Citizen	-0.015*	0.007
Immig Non-Citizen	-0.015**	-0.003
Father
High School Education	-0.013**	-0.005
College Education	-0.027***	-0.023
Mother
High School Education	-0.016***	-0.012**
College Education	-0.035***	-0.009

Slide 16

Discrete Hazard Results: Parent Stay - 1 Parent HH

	Citizen Immigrant Parent (&)		Non-Citizen Immigrant Parent (&)
	Entry	Retention	Entry	Retention
<5 years	0.048	0.082	0.019	0.020*
5-10 years	0.049	0.083*	0.011	0.012
10-15 years	0.003	0.027	0.007	0.009
>15 years	0.003	0.026	-0.002	0.001

Comparison group: Native Parent.

Slide 17

Steady State (Probability Enrolled) by Parent's Stay: 2 Parent HH

	Citizen Immigrant Father	Non-Citizen Immigrant Father	Citizen Immigrant Mother	Non-Citizen Immigrant Mother
<5 years	-0.007	-0.001	0.006	0.003
5-10 years	-0.006	-0.001	0.005	0.001
10-15 years	-0.009**	-0.004	0.004	-0.001
>15 years	-0.009**	-0.003	-0.003	0.008

Slide 18

Steady State (Probability Enrolled) by Parent Stay: 1 Parent HH

	Citizen Immigrant Parent	Non-Citizen Immigrant Parent
<5 years	0.002	0.007*
5-10 years	0.001	0.006
10-15 years	-0.001	0.005**
>15 years	-0.001	0.006*

Slide 19

Conclusions and Policy Implications

• Overall, participation rates among children of immigrant and native born parents are very similar.
• Differences depend on household structure and nativity status of a mother or father
  • Hazard of Entry into public coverage.
    • Decreases for children with citizen immigrant father.
    • Increases for children with non-citizen immigrant mother.
  • Hazard of Retaining public coverage
    • Does not depend on the nativity status of child's father.
    • Decreases for children with non-citizen immigrant mother.
• For one parent households, children with a foreign born parent exhibit assimilation out of public coverage.

Slide 20

Conclusions and Policy Implications

• Higher parental education and being an immigrant child affect entry into and retention of public coverage
  • Education - attitudes to public porgrams and/or job opportunities?
  • Foreign born child - Chilling effect?
  • Overall, results are robust to alternative specifications.
• No evidence of assimilation into public coverage among children with foreign born parents
  • Contrary to the literature on adult immigrants' participation in welfare programs.