All in the Family? Children's Public Coverage, Dual-Earner Households and Employer-Sponsored Insurance

Select to access the PowerPoint® presentation (265 KB).

Slide 1

All in the Family? Children's Public Coverage, Dual-Earner Households and Employer-Sponsored Insurance

Jessica Vistnes
(AHRQ)

Kosali Simon
(Indiana University, SPEA and NBER)

Slide 2

Motivation

• Dependent health insurance for workers' spouses and children is an important source of coverage in the U.S.
• Roughly half of non-elderly Americans covered by employer-sponsored insurance have that coverage as a dependent (Current Population Survey, 2008).

Slide 3

Average Employer-Sponsored Health Insurance Premiums: 2011

Source: 2011 Medical Expenditure Panel Survey—Insurance Component, Private Sector Establishments.

Slide 4

Employees' Enrollment Decisions

• Two-worker Couples with Two Offers of Insurance:
• Without children:
  • 2 single policies vs. 1 family policy?
  • Employee-plus-one coverage, if available?
• With children:
  • Are eligible for public coverage?
  • Different sources of coverage to cover everyone?

Slide 5

Employers' Decisions

• Offer coverage at all?
• Offer dependent coverage?
• How to set employee contributions?
• Restrict access to coverage for worker's spouse?
• Offer cash incentives if workers decline coverage?

Slide 6

Prior Literature

On effects of coverage through a working spouse:

• Dranove, Spier and Baker (2001).
• Vistnes, Morrisey, Jensen (2006).

On effects of public coverage for children:

• Shore-Sheppard, Buchmueller, Jensen (2000).
• Buchmueller, Cooper, Simon and Vistnes (2005).

Slide 7

Changes in Dependent Coverage 2000-2008

• Dependent coverage is less likely to be offered at small employers.
• Employee-plus-one coverage more likely to be offered.
• Employee premium contributions have risen.

Slide 8

Availability of Dependent Coverage

Images: Two bar graphs show availability of dependent coverage for firms with less than 10 workers and for all firms.

Firms with less than 10 workers (Any Dependent Coverage Offered):

• 2000 - 89.4.
• 2001 - 87.6.
• 2008 - 82.6.

All Firms (Employee-Plus-One Coverage Offered):

• 2000 - 70.1.
• 2001 - 78.0.
• 2008 - 87.4.

Slide 9

Goals

• To re-examine the effects of alternative sources of coverage on a number of ESI outcomes.
• To take into account rising unemployment rates over our analysis period and investigate whether the effects of alternative sources differ at low and high unemployment rates.

Slide 10

MEPS-Insurance Component (MEPS-IC)

• 2005-2010 MEPS-IC:
  • Large, annual, nationally representative establishment level survey.
  • Sponsored by AHRQ, conducted by the Bureau of the Census.
• Collects information on:
  • Offers of insurance, establishment/workforce characteristics.
  • Number and types of plans.
  • Total premiums, employee and employer contributions.
  • Deductibles, copayments/coinsurance and other benefit details.

Slide 11

Data

• Information on full-time workers' family characteristics:
  • Estimates calculated from the 2005-2010 American Community Survey (ACS).
  • Merge onto the MEPS-IC by state, detailed industry and year.
• 2005-2010 Area Resource File:
  • County Unemployment Rate.
  • Other variables.
• Estimate models separately by firm size
  Small : <100 and Large: ≥100 workers.

Slide 12

Dependent Variables: Establishment Level

• Offers:
  • Any coverage.
  • Any dependent coverage.
• Take-up rates.
• Enrollment Shares for single, employee-plus-one, family.
• Restrictions on spousal coverage.
• Financial incentives to decline coverage.

Slide 13

Dependent Variables: Plan Level

Slide 14

Key Independent Variables

• Proportion of full-time workers (From ACS):
  • Married.
  • Have children.
  • Have children and are <200% FPL.
  • In families with two full-time working spouses.
• Eligibility index for Medicaid/CHIP coverage for children.
• Unemployment rates.
• Interaction terms:
  • In families with two full-time working spouses * female.
  • Have children and are <200% FPL * Medicaid/CHIP Eligibility.

Slide 15

Other Explanatory Variables

• Establishment Characteristics:
  • Firm size, industry, age of business, ownership type, non-profit status.
• Workforce Characteristics:
  • Proportion of workers that are—
    • Age 50 and older, female, union members.
  • Wage distribution (three wage categories).
• Plan type, # plans, single premium in plan equations.
• State and year fixed effects.
• County level characteristics from Area Resource File.

Slide 16

Outcome: Offer and Take-Up
Small Firms

Slide 17

Offer model with unemployment interactions
Small firms

Estimated a second model that added the following two variables:

Unemployment rate * two-worker-family
Unemployment rate * two-worker-family * female

Slide 18

Outcome: Offer Equation with Unemployment Interactions—Small Firms

What if the % Two-Worker increases from 0 to 30%?

• At establishments that are 50% female.

Slide 19

Outcome: Family—Single Contributions
Large Firms

Increasing % Two-Worker:

• At 50% female.
• Moving from 0% to 30% for two-worker families:
  • $508 increase in marginal family contributions.

Increasing % Low-Income Children:

• At mean value for public eligibility.
• Increasing proportion with children <200% FPL: 0 to mean values.
• $142 increase in marginal family contributions.

Slide 20

Outcome: Spousal Restriction on Coverage
Large Firms

• What if the percent of "workers married" increases from 0 to 50%?
• And the percent of "2-worker households" increases from 0 to 25%?

Image: A line graph captioned "Percentage Point Change in Spousal Restriction on Coverage" shows the percentage point change decrease dramatically as the percentage of females increases.

Slide 21

Outcome: Family enrollment shares
Large firms

• What if we increase "% married" from 0 to 50%?
  • 10 percentage point increase in family share.
• What if we increase "% with children" from 0 to 50?
  • 10 percentage point increase in family share.

Slide 22

Outcome: Enrollment Shares
Large firms

What if we increase the "% in two-worker families"?

• Family shares go down [Image: An arrow points down].
• Single and employee-plus-one shares (p <.10) go up [Image: An arrow points up].

Slide 23

Outcome: Cash Incentives

• Large firms:
  • More likely to offer cash incentive as proportion with children increases.
    • May reflect that these employees are more likely to enroll in expensive family coverage.
  • Less likely to offer cash incentive as the proportion in two-worker families increases.
    • May reflect that these employees are less likely to enroll in family coverage.
• Small firms:
  • Positive, significant effects on variables associated with alternative sources of coverage for children and adults.

Slide 24

Summary

• Alternative sources of coverage affect:
  • Offers of coverage.
  • Restrictions/incentives by employers.
  • Premium contributions.
  • Enrollment decisions.
• No distinction can be made between:
  • Employers acting as agents for workers.
  • Strategically encouraging alternative coverage.

Slide 25

The Affordable Care Act

• The ACA introduced new alternative sources of coverage and new incentives for workers and employers.
• Understanding how dependent coverage has changed in recent years helps set the context for changes that will occur under the ACA.