Special Analysis

Implications for Family Planning of Post-Welfare Reform Insurance Trends

Rachel Benson Gold, Guttmacher Institute
Reproductive rights are under attack. Will you help us fight back with facts?

First published online:

New tabulations by The Alan Guttmacher Institute of Census Bureau data show that between 1994 and 1998, the proportion of American women of reproductive age enrolled in Medicaid fell by 21%. By 1998, nearly one in five American women of reproductive age lacked insurance of any kind. While the precise causes and effects are not yet completely clear, welfare reform is likely to have played a major role in these developments, which have important implications both for women needing family planning services and for the health care professionals seeking to serve them.

Historically, eligibility for Medicaid—the joint federal-state health insurance program for very poor Americans—was tied to eligibility for cash assistance under the Aid to Families with Dependent Children program, commonly known as public assistance or welfare. Families who met the eligibility requirements for welfare were automatically enrolled in Medicaid; in most cases, there were neither separate requirements nor a discrete Medicaid enrollment process.

With passage of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996—welfare reform—Congress ended the decades-old entitlement to public assistance, instead placing a time limit on a family's eligibility for aid. As the legislation took shape, health care advocates were concerned that as a result of losing welfare benefits, large numbers of families would automatically lose their Medicaid coverage as well.

Responding to these concerns, Congress included a provision aimed at holding Medicaid eligibility constant, despite the dramatic changes being made to welfare. Specifically, an amendment added largely at the behest of the late Sen. John Chafee (R-RI) provided that families meeting what had been a state's eligibility requirements for Medicaid prior to the legislation would be eligible in the future for Medicaid coverage regardless of whether they were eligible for welfare. This provision, which effectively delinked eligibility for Medicaid from eligibility for welfare, was designed both to ensure that families already enrolled in Medicaid would continue to be covered and to permit additional families to enroll in the program even if they did not meet the new requirements for welfare.

Implementation Hurdles

In the mid-1980s, Congress partially delinked Medicaid and welfare eligibility when it expanded Medicaid to cover pregnant women and young children solely on the basis of their income, regardless of whether they received welfare. While the program scored impressive gains in enrolling pregnant women, at least by the time of delivery—likely because these women easily could be enrolled when they entered the hospital to give birth—success was more limited with young children who had no immediate need to interface with the health care system. According to researchers at The Urban Institute, nine in 10 children in 1993 who were eligible for Medicaid because of their receipt of welfare benefits were actually enrolled in the program, compared with only seven in 10 who were eligible solely because of their income. This experience clearly foreshadowed the problems facing the states in ensuring Medicaid coverage in the wake of welfare reform.

To fully carry out the intent of the Chafee amendment, state Medicaid programs face three key challenges. First, they need to retain Medicaid coverage for families losing cash assistance as a result of the limitations imposed by the welfare reform legislation. But according to Leighton Ku, an expert on welfare and Medicaid policy at The Urban Institute, this is by no means as easy as it might appear. "It's very difficult," Ku says, "to communicate that you're being terminated from welfare but that you remain eligible for Medicaid, especially because the long-standing link between the two programs likely continues in the minds of both many recipients and eligibility staff." To help with this problem, the legislation extended the life of a small "transitional" Medicaid program, designed to assist families moving from welfare to work by automatically continuing their Medicaid coverage for a period of time. Unfortunately, according to the U.S. General Accounting Office (GAO), this program is performing below expectations, in part because, as Ku suggests, people typically do not realize that they can retain Medicaid coverage even if welfare coverage is lost and because of an array of cumbersome income-reporting requirements.

The second major challenge facing states is to ensure that welfare policies and restrictions do not pose a barrier to new families seeking Medicaid coverage. To do so, states in effect would have to create Medicaid eligibility and enrollment procedures from scratch. Instead, according to the Center for Budget and Policy Priorities, all states appear to have retained a joint application procedure for welfare and Medicaid, even though the eligibility criteria now differ for the two programs. As a result, states need to find ways to make sure that any preconditions for welfare eligibility—such as a requirement that applicants demonstrate that they are actively seeking work—do not impede processing of the Medicaid component of the application. This task is further complicated, according to GAO, by the fact that eligibility workers previously handled very few discrete applications for Medicaid, and they may not be completely comfortable with the program's complex and seemingly arcane eligibility rules.

Third, states must develop programs to actually reach out to eligible families that may not initiate contact with the system on their own, out of ignorance, fear or the lingering stigma of the program's past association with welfare. To help states grapple with this issue, the 1996 legislation established a $500 million outreach fund, upon which states could draw to conduct a variety of public education efforts, including establishing toll-free hot lines, conducting media campaigns and distributing informational literature and other materials.

As the outreach fund was initially designed, states would have three years from the initiation of their welfare reform efforts to spend their allocations under the fund. This means that the eligibility of 33 states was slated to terminate by the end of 1999, with the entire fund essentially going out of business by October 1, 2000. Despite these looming deadlines, according to the Health Care Financing Administration (HCFA), the federal agency that administers Medicaid, spending under the fund has been minimal; states had submitted claims for only $50 million of the $500 million allotment by the middle of 1999.

The Special Case of Immigrants

While the welfare reform legislation intended to maintain Medicaid eligibility for most current and potential enrollees, it singled out immigrants, both those in the country illegally and many of those here legally, to lose coverage altogether. Under the new law, immigrants living in the country illegally are simply ineligible for Medicaid. The ability of legal immigrants to obtain Medicaid depends on when they arrived in the United States. Immigrants who came before August 22, 1996, the date the legislation was enacted, continue to be eligible. But those who came after that date are ineligible for five years after their arrival. (Many advocates argue that coverage is unlikely even then because of a related provision that will allow states to take into account both the immigrant's income and the income of the immigrant's sponsor for purposes of determining Medicaid eligibility.)

As with the delinking of Medicaid and welfare, the impact of the limitations on immigrants is a combination of the policy itself and its implementation. Many advocates contend that as important as the actual restrictions themselves is the climate of fear and distrust that the legislation created. According to the National Health Law Program, many immigrants may be unwilling even to apply for Medicaid coverage to which they may be entitled out of a concern that to do so will somehow jeopardize their immigration status or that of family members, despite recent attempts by the Immigration and Naturalization Service to make clear that these fears are unfounded.

Impact on Coverage

At a time when the economy is booming and unemployment is extremely low, it might be expected that more and more people would have employment-related health insurance. But new data indicate that in the midst of an extraordinary period of economic expansion, the number of Americans lacking health insurance is actually on the rise. While it is still too early to precisely define the role of welfare reform in the rise in the number of uninsured Americans or the decrease in the number covered by Medicaid, The Urban Institute's Ku says that most observers believe welfare reform to be a principal contributor to these trends.

Newly released data from the Current Population Survey, which is conducted by the Census Bureau, show that in 1998, 16.3% of all Americans (44.3 million Americans) lacked any insurance coverage whatsoever, compared with 15.2% in 1994, an increase of 7%.

New tabulations by The Alan Guttmacher Institute of these survey data on the insurance coverage of women of reproductive age (15-44) mirror this national trend. Nationally, 12 million women of reproductive age were uninsured in 1998, 1.5 million more than had been uninsured in 1994. The proportion of women of reproductive age who were uninsured rose from 17.5% in 1994 to 19.8% in 1998, an increase of 13% (see chart). By 1997-1998, more than one in five women of reproductive age in 18 states were uninsured (see table). The proportion of women who were uninsured rose in 37 states over this period; significant increases occurred in 15 states.

Coverage Declines
The proportion of women of reproductive age who are uninsured is rising, as the proportion covered by Medicaid is falling.
Figure 1
% of women 15-44
Percentage of Women 15-44 Uninsured or Covered by Medicaid, 1994-1995 and 1997-1998
State % Uninsured % with Medicaid
1994-1995 1997-1998 1994-1995 1997-1998
Alabama 20.6 22.4 9.7 7.9
Alaska 13.7 17.4* 14.4 9.7*
Arizona 25.1 27.7 11.2 9.1
Arkansas 22.0 25.2 11.0 10.1
California 22.9 26.4* 16.6 12.5*
Colorado 15.1 18.8* 6.4 3.2*
Connecticut 10.4 17.4* 9.5 7.0
Delaware 18.0 18.4 8.7 9.5
Dist. of Columbia 18.1 17.1 23.1 19.2*
Florida 21.9 22.6 11.4 8.3*
Georgia 19.8 22.1 10.8 12.0
Hawaii 10.6 8.7 12.3 11.0
Idaho 16.1 22.4* 11.5 7.6*
Illinois 12.9 15.8* 12.8 9.3*
Indiana 14.3 16.8 8.1 4.2*
Iowa 12.6 11.7 8.1 8.5
Kansas 15.5 14.7 8.6 9.2
Kentucky 18.2 18.6 15.3 11.7*
Louisiana 24.2 23.4 15.8 12.3*
Maine 17.3 18.2 10.6 7.6*
Maryland 16.3 16.4 9.6 4.2*
Massachusetts 13.8 12.8 11.6 13.3
Michigan 12.0 17.0* 14.6 13.7
Minnesota 11.1 10.2 12.8 11.7
Mississippi 24.2 25.0 17.9 8.0*
Missouri 16.8 13.4* 10.3 9.2
Montana 17.4 23.7* 14.3 9.6*
Nebraska 12.3 11.6 6.3 8.9*
Nevada 17.3 21.3* 6.0 5.1
New Hampshire 13.6 12.6 8.5 8.8
New Jersey 16.0 19.2* 8.5 6.4*
New Mexico 30.1 28.5 16.1 12.3*
New York 18.9 21.1* 16.2 15.1
North Carolina 16.1 18.2 11.6 9.3*
North Dakota 9.8 17.1* 7.8 6.8
Ohio 14.2 12.8 12.2 9.2*
Oklahoma 21.8 22.3 14.9 6.5*
Oregon 14.6 16.1 17.4 15.9
Pennsylvania 12.1 12.9 12.3 11.2
Rhode Island 16.1 12.1* 10.2 10.8
South Carolina 15.3 21.8* 11.5 8.1*
South Dakota 10.5 16.8* 11.9 8.7*
Tennessee 15.5 17.0 17.7 21.6*
Texas 27.3 27.9 10.3 8.3*
Utah 12.9 15.6* 6.3 6.6
Vermont 13.2 10.6 9.9 17.3*
Virginia 14.6 16.0 6.2 5.9
Washington 15.4 15.9 14.9 11.4*
West Virginia 22.4 24.0 14.7 14.7
Wisconsin 9.4 11.0 8.7 7.7
Wyoming 19.8 21.3 7.6 7.1
*Difference is significant at p<.10. Source: March Supplements of the Current Population Survey for four years: 1994, 1995, 1997, 1998. This nationally representative sample survey of approximately 50,000 households is carried out annually by the Bureau of the Census. Notes: Two years of survey data were combined to obtain state-level estimates. This is necessary to increase sample size, reduce sample error and increase the stability of the estimates. Nevertheless, in interpreting differences between states and over time, it must be borne in mind that, as survey-based estimates, the data presented are subject to sampling and nonsampling error. For further details, see the following Web site: <"http://www.bls.census.gov/cps/ads/1999/ssrcacc.htm">www.bls.census.gov/cps/ads/1999/ssrcacc.htm&gt;.

Experts believe that much of the increase in the number of uninsured Americans is due to a drop in Medicaid coverage among Americans who are neither elderly nor disabled. According to GAO, Medicaid enrollment nationwide dropped by 1.7 million, or 7%, between 1995 and 1997.

Again, the data on women of reproductive age mirror the national trend. In 1998, Medicaid covered 6 million women in that age-group, 1.5 million fewer than the program had covered in 1994. The proportion of women of reproductive age covered by Medicaid fell from 12.6% in 1994 to only 9.9% in 1998, a 21% decrease (see chart). The proportion of women covered by Medicaid fell in 38 states and the District of Columbia; drops were significant in 22 of these states and the District of Columbia (see table).

According to data from The Urban Institute, coverage for low-income noncitizens, the only group intentionally dropped from the Medicaid rolls by welfare reform, follows a similar trend. The Urban Institute estimates that between 1995 and 1998, the proportion of low-income immigrants without insurance rose from 54% to 59%, while the proportion covered by Medicaid fell from 19% to 14%.

Implications for Family Planning

Family planning providers are feeling the effects of these trends. The number of Medicaid recipients served by the Family Planning Council of Southeastern Pennsylvania, for example, fell by 5,000 between 1996 to 1999—representing a loss to the council of almost $1 million in Medicaid revenues—while the number of uninsured clients rose by 4,000. According to Executive Director Dorothy Mann, the deficit caused by this lost Medicaid revenue has seriously hampered the council's ability to accommodate recent growth in the overall number of clients seeking services from the agency, leading in turn to longer waits to schedule appointments, more crowded clinics and longer waiting time for patients in the clinics.

Meanwhile, funds available to serve patients without insurance are increasingly strained. Peter Durkin, chief executive officer of Planned Parenthood of Houston and Southeast Texas, says his agency has run out of its Social Services Block Grant (SSBG) allotment in midyear in recent years. "We depend on the SSBG to serve uninsured women," Durkin says. "With Congress having just slashed funding for that program by nearly $200 million for this year, it is going to become even harder to find a way to serve the increasing number of uninsured women who are seeking care."

One particularly troubling consequence of the welfare reform law for family planning providers stems from the fact that immigrants, while not eligible for regular Medicaid coverage, can receive coverage for emergency medical care. As a result, a woman in labor may be covered by Medicaid for delivery but not for postpartum family planning. Karen Gluck of the University of Texas Southwestern Medical School at Dallas contends that such a policy is illogical. "To take people who are cut off from welfare," Gluck says, "and give them medical care only during childbirth but not the services they need to avoid a subsequent pregnancy if they desire to do so makes no sense at all." The result, according to Gluck, is a "scramble for funds from some other source. We can't just let them go without care."

Tackling the Problems

These early indications of the spillover effects of welfare reform on Medicaid coverage are drawing widespread attention from advocates and policymakers alike. One of the first targets of this attention is likely to be the transitional Medicaid program. The Clinton administration is supporting legislation to simplify reporting requirements in hopes of making the program a more viable source of coverage for former welfare recipients. Meanwhile, HCFA in March released a guide to the states that provides an array of strategies to simplify and improve this effort.

The administration this year also launched a multifaceted effort to ensure that eligible individuals are informed about Medicaid eligibility requirements and to facilitate their enrollment. In August, President Bill Clinton instructed HCFA to conduct on-site reviews of state Medicaid programs "to ensure that there are no roadblocks, intentional or, even more likely, unintentional, to those who are eligible for Medicaid." A key component of this campaign is to address GAO's finding that some states are having trouble communicating "to both beneficiaries and [eligibility] workers that Medicaid coverage can be maintained" even if welfare payments are terminated.

To provide a source of funding for these activities at the state level, both the administration and advocates worked to prevent the sunset of the welfare-Medicaid outreach fund. These efforts proved successful, and an extension of the fund was included in the final budget compromise worked out between the administration and Congress. This will give states additional time to spend much of the remaining $450 million that was originally allocated to this effort.

Finally, several members of Congress have proposed loosening the limitations on Medicaid coverage for at least some groups of immigrants. In 1997, Congress restored Medicaid coverage to some disabled immigrants. And bills are now pending in the House and Senate to restore Medicaid coverage to pregnant immigrants and their children. For the women, this would end the current policy of providing only emergency treatment at the point of delivery, but not the prenatal and postpartum care—including family planning—that is important to reduce both the likelihood of adverse pregnancy outcomes and the likelihood of a subsequent unintended pregnancy.


Tabulations of data from the Current Population Survey were done by Rachel K. Jones, senior research associate, The Alan Guttmacher Institute. The research on which this article is based was supported in part by the General Service Foundation and the U.S. Department of Health and Human Services (grant FPR000057). The conclusions and opinions expressed in this article, however, are those of the author and The Alan Guttmacher Institute.