A Lifetime of Punishment: The Impact of the Felony Drug Ban on Welfare Benefits

In his first State of the Union address, President Bill Clinton promised to “end welfare as we know it.”¹ Nearly four years later, on August 22, 1996, President Clinton signed legislation to do exactly that: the Personal Responsibility and Work Opportunity Reconciliation Act (PRWORA).²

PRWORA’s reforms were expansive and controversial for several reasons, including its implementation of a revised cash assistance program—Temporary Assistance to Needy Families (TANF)—which limited the length of time eligible families could receive benefits³ and established work requirements for recipients.⁴ In addition, PRWORA made substantial changes to the operation of the federal food stamp program,⁵ which has since been renamed the Supplemental Nutrition Assistance Program (SNAP).

Perhaps because of the general debate surrounding PRWORA’s changes to cash assistance and food stamp programs, one significant provision of the law initially received little attention: along with other federal legislation related to the “war on drugs,” PRWORA imposed a denial of federal benefits to people convicted in state or federal courts of felony drug offenses. The ban is imposed for no other offenses but drug crimes. Its provisions that subject individuals who are otherwise eligible for receipt of SNAP or TANF benefits to a lifetime disqualification applies to all states unless they act to opt out of the ban.⁶

Despite the magnitude of this change, the provision received only two minutes of debate after it was introduced on the Senate floor—one minute for Republicans and one minute for Democrats.⁷ It was then unanimously adopted by a voice vote.⁸ The brevity of Congressional discussion on the felony drug conviction ban makes it difficult to know the intent of Congress in adopting this policy, but the record that does exist suggests the provision was intended to be punitive and “tough on crime.” As Senator Phil Gramm (R-TX), the sponsor of the amendment, argued, “if we are serious about our drug laws, we ought not to give people welfare benefits who are violating the Nation’s drug laws.”⁷ Conspicuously absent from the brief debate over this provision was any discussion of whether the lifetime ban for individuals with felony drug offenses would advance the general objectives of welfare reform.

In an effort to assess the impact of this policy, this report provides an analysis of the ban on receipt of TANF benefits for individuals with felony drug convictions. First, we survey the current status of the ban at the state level, including actions by legislatures to opt out of the ban in full or in part. Next, we produce estimates of the number of women potentially affected by the ban in those states that apply it in full. We then assess the rationale for the ban and conclude that, for a multiplicity of reasons, the ban not only fails to accomplish its putative goals, but also is likely to negatively impact public health and safety. Finally, we offer policy recommendations for future treatment of the ban on receipt of food stamps and cash assistance for individuals convicted of felony drug crimes.

State Policies

Although PRWORA banned the receipt of SNAP and TANF benefits for individuals with felony drug convictions, it gave states the discretion to opt-out of or modify the ban. By 2001, eight states and the District of Columbia had entirely opted out of the ban, while an additional 20 states had modified it.¹⁰ In the last decade, more states have joined the ranks of those that do not enforce PRWORA’s drug-crime exclusion provisions in full.

Despite these changes, a 2011 review of state policies by the Legal Action Center documents that three-quarters of the states enforce the ban in full or in part.¹¹ Currently, 37 states either fully or partially enforce the TANF ban, while 34 states either fully or partially enforce the SNAP ban (Table 1). Of these states, 25 (largely, but not precisely the same for both policies) have modified the ban to allow individuals with felony drug convictions to receive TANF or SNAP benefits under certain circumstances. For example, Arkansas, Florida, and North Dakota allow people to receive TANF if they were convicted of possessing drugs, but not manufacturing or distributing drugs. Other states allow receipt of TANF benefits for individuals who take part in or complete drug treatment, submit to drug testing, or have completed a specified waiting period. North Carolina, for instance, bans people from receiving benefits for six months following completion of a felony drug sentence. Although states are minimally more lenient in allowing people to receive food stamps, SNAP restrictions generally mirror state TANF restrictions.

a As of 2015, the states that fully ban persons with felony drug convictions from receiving TANF benefits include Arizona, Michigan, and Virginia
b Source: http://bit.ly/HIRE_TANF
c Source: http://www.lac.org/toolkits/TANF/TANF.htm#summary

Impact of the Federal Ban on TANF

The federal ban on TANF benefits has been in effect since 1996. Given the scale of drug convictions annually, the number of individuals affected by the ban is potentially quite substantial. In this analysis we develop estimates of this effect. To produce a conservative estimate of the impact of the ban, we use the following methodology:

• First, since state policies vary somewhat between prohibitions on TANF or SNAP we focus here only on the TANF ban. We do so because the financial effect of the TANF ban is more significant for affected households, but with the recognition that many of the individuals excluded under the TANF ban have also lost food stamp benefits.
• Our analysis only covers the 12 states that impose a full ban on TANF benefits. Although there are an additional 25 states that impose a partial ban, there is no reliable means of obtaining data on the factors that trigger these bans (such as distinctions between convictions for drug sales or drug use, or the number of people with felony drug convictions enrolled in treatment programs).
• Our analysis only covers the effect on women with felony drug convictions. Although the absolute number of men with drug convictions is far greater, women with children are far more representative of the TANF population.

Our estimates represent the lifetime potential impact of the TANF ban in these selected states. That is, the prospect that at some point in their lives women who would otherwise qualify for such benefits will be denied them due to a prior felony drug conviction.

At any given moment in time, many women would not qualify for these benefits since eligibility criteria include having custody of minor children, meeting income and work requirements, and not having exhausted the lifetime eligibility limit (five years in most states).¹² Eligibility for food stamps is similar, except non-parents are also eligible to receive SNAP benefits.¹³

The estimated number of women potentially affected by the PRWORA ban in states that fully ban people convicted of drug felonies from receiving TANF was derived using data from the Bureau of Justice Statistics’ Felony Sentences in State Courts series.¹⁴ These data are based on a nationally representative sample of counties and are available for even years from 1996 through 2006. The average of the preceding and subsequent years was used to estimate odd year values, and the 2006 value was used to approximate values for each year from 2007 through 2012.

Estimates of the proportion of sentencing events involving women and the proportion in which a drug offense was the most serious offense were used to estimate the number of women convicted annually of a felony drug offense. The annual estimates were adjusted downward to account for multiple convictions during one sentencing event. Further adjustments were made to account for mortality and reconviction over time. These adjustments led to an estimate of the unique number of women convicted of felony drug offenses from 1996-2011. The estimated total number of women convicted of drug felonies from 1996 through 2011 was apportioned to states according to the estimated proportion of the national combined female probation, parole, and prison populations within those states.

As seen in Table 2, for the 15-year period 1996 – 2011 there are now an estimated 180,100 women in these states who may be affected by the TANF ban at some point in their lives.

¹ The Bureau of Justice Statistics reports that probation counts in Georgia may overstate the number of individuals under supervision because the agency that reports the county data has the capacity to report probation cases, but not the number of individuals under supervision. Therefore, individuals on probation with multiple sentences may be under supervision by more than one agency.

Note that the number of individuals affected would greatly increase if the analysis were expanded to include women in the 25 states that partially implement the ban or who are only seeking SNAP benefits, as well as low-income men with felony drug convictions.

The Ban’s Disparate Effects

While the TANF ban does not target any demographic groups specifically, the dynamics of social class and the accompanying disparate racial effects of criminal justice policy and practice combine to produce highly disparate effects on women, children, and communities of color.

Impact on Women

The ban’s effect on women results from several factors. First, women comprise the vast majority of recipients of both TANF and SNAP benefits. In 2009, 85.9% of adult TANF recipients were women;¹⁵ women are also about twice as likely as men to receive food stamp benefits at some point in their lives.¹⁶

Law enforcement and sentencing trends in recent decades have also combined to skew the effect of the ban on women. This has come about through two interrelated trends – a sharply rising number of women charged with drug offenses and a disproportionate effect of drug law enforcement on women. While prison populations have grown dramatically in recent decades, the rise in women’s incarceration has outstripped that of men. From 1980 to 2010, the number of women in prison rose by 646%, compared to a 419% increase for men.¹⁷

From 1980 to 2010, the number of women in prison rose by 646%, compared to a 419% increase for men.

Within the prison population, women have been affected more so than men by drug law enforcement. Given that women are typically a small percentage of people who commit violent crimes, their numbers in prison historically were quite low. But as drug law enforcement accelerated rapidly beginning in the 1980s, women became much more likely to be convicted of a felony or sentenced to prison than in previous eras. By 2011, 25.1% of women in state prisons were incarcerated for a drug offense, compared to 16.2% of men.¹⁸ Thus, the combination of the high rate of women as SNAP and TANF recipients, along with the disproportionate effect of the drug war on women, has produced the skewed effects of the PRWORA ban.

Impact on Children

In addition to the direct effect of the TANF ban on parents, the ban also has an immediate impact on their children, who have committed no crime themselves. Under the terms of the law, in a TANF-eligible household the monthly grant allotment is reduced for the ineligible parent, but is still allowed for that person’s children. For example, if a single mother with two dependent children has a felony drug conviction the TANF benefit will be reduced from the three-person level to that of a two-person household. Given that TANF benefits are quite modest to begin with, a reduction of this size creates substantial additional hardship for such families.

Racial / Ethnic Impact

The federal ban on receipt of food stamps and cash assistance for individuals with felony drug convictions disproportionately impacts African Americans and other minority groups. This is a direct reflection of the racial disparities produced by the “war on drugs.” Data on illicit drug use collected by the Department of Health and Human Services has consistently shown over time that whites, African Americans, and Latinos use drugs at roughly comparable rates.¹⁹ But as of 2011, African Americans comprised 40.7% of prisoners in state prisons for drug crimes, while individuals of Hispanic origin made up another 21.1% of this population.²⁰ Thus, the racial/ethnic disparities in drug offender incarceration produced by the interaction of law enforcement and sentencing policies through the war on drugs then translate into a disproportionate impact of the felony drug ban.

Assessing the Ban as Policy

As we have seen, the felony drug ban potentially affects hundreds of thousands of women (as well as children and men) over the course of their lifetimes, well after most will have completed serving their felony sentences. For this disproportionately lower-income population, the sudden loss of a job or change in family circumstances can move an otherwise self-supporting household into a situation whereby the loss of federal benefits can make the difference between stability and vulnerability in one’s life prospects.

In order to justify such effects, we can explore the possible beneficial effects of the ban that may have motivated federal lawmakers to adopt the policy originally, and to determine to what extent the policy of benefits denial has succeeded in its goals. Although members of Congress did not specifically articulate a rationale for the ban, it has often been assumed that denying SNAP and TANF benefits to individuals convicted of drug crimes arose out of “the government’s desire to deter drug use and to reduce incidences of fraud.”²¹ The following is an assessment of the ban’s effect on these goals, which leads us to conclude that the ban is not necessary to or effective at achieving them.

Deterring Drug Use

To the extent that policymakers believed that the ban on benefits would deter use, they were unfortunately very misinformed about the connection between substance abuse and certain criminal behaviors. While the ban applies to individuals convicted of a drug offense, many people in this category do not use drugs themselves. Looking at data from 2006 (most recent available) from the Bureau of Justice Statistics, we find that more than half (56%)²² of the 377,860 drug convictions that year were for selling drugs, not using drugs. Some people who sell drugs do so to support their own drug use or addiction, but many do so as a means of making money. In addition, of the remaining 44% of drug convictions for possession, many were for the offense of “possession with intent to deliver,” a charge involving sale of drugs. Therefore, the welfare ban applies to many people convicted of a drug crime who do not use drugs, but does not apply to drug users who have been convicted of larceny, theft, robbery, and a host of other felonies.

Denying individuals convicted of drug crimes food stamps and cash assistance is one of the many collateral consequences of a felony conviction that have been termed an “invisible punishment”—a sanction that results from a criminal conviction but “take[s] effect outside of the traditional sentencing framework,” and as a result “operate[s] largely beyond public view, yet ha[s] very serious, adverse consequences for the individuals affected.”²³ Collateral consequences in general have dubious value as deterrents, in large part because most people are unaware of the civil penalties that result from criminal convictions.

There is little reason to believe that barring individuals with felony drug convictions from receiving welfare benefits deters drug use or crime.

In particular, there is little reason to believe that barring individuals with felony drug convictions from receiving welfare benefits deters drug use or crime. For example, one study of women with drug convictions or pending felony drug charges found that not a single one of the 26 women interviewed was aware prior to her involvement with the criminal justice system that a felony drug conviction could lead to a loss in SNAP or TANF benefits.²⁴ Furthermore, 92% of the women reported that even if they had known of the ban, it “would not have acted as a deterrent during active addiction.”²⁵ Because of the nature of addiction, it is also generally implausible to believe that a person who is not deterred from criminal activity by the specter of criminal prosecution or imprisonment would be halted by the threat of losing access to TANF and SNAP benefits.

Reducing Welfare Fraud

The ban on receipt of TANF and SNAP benefits for individuals with felony drug convictions is sometimes defended on the ground that the ban helps to reduce fraud in the federal welfare system. The logic of this claim seems to be that individuals with drug convictions are more likely to be drug users, and that drug users are more likely to commit welfare fraud—for example, by using TANF cash payments to buy drugs or by trafficking food stamps.²⁶

The perception that drug users may be likely to commit fraud may be traceable, in part, to “[a] series of media accounts in the early 1990s,” which “suggested that food-stamp benefits were being exchanged readily for cash and contraband.”²⁷ Scholars have noted that the problem with these accounts is that they often involved undercover officers who tried to exchange food stamps for cash, drugs, or weapons, and that while their success in doing so demonstrates that food stamps have value, “[t]hese anecdotes did not establish that households receiving monthly food-stamp allotments—as opposed to undercover agents with benefits provided explicitly for sting operations—were exchanging food stamps improperly.”²⁸ In reality, the SNAP fraud rate is extremely low: from 2006-2008, the trafficking rate for food stamps was approximately one cent per every dollar.²⁶ At least one explanation for the low fraud rate is the fact that SNAP benefits are now issued on an electronic benefit card that functions like a regular debit card and makes it both harder to misuse benefits and easier for the government to identify and track suspicious food stamp activity.²⁶

Even though the fraud rate is low, it is not unreasonable to attempt to detect and prevent the trafficking of food stamps. But disallowing TANF and SNAP benefits to individuals with felony drug convictions is hardly necessary to achieve this goal since federal legislation already proscribes and punishes fraudulent use of welfare benefits.³¹ In fact, trading controlled substances for SNAP benefits is specifically prohibited in a separate section of the United States Code; individuals who are found to have traded controlled substances for SNAP benefits are punished with two years of SNAP ineligibility for a first offense and permanent ineligibility for a second offense.³¹ This provision is more closely tailored to the purpose of deterring food stamp fraud than the blanket ban on receipt of food stamps for individuals with felony drug convictions, because it is responsive to actual misuse of benefits regardless of whether the recipient has a history of criminal or drug involvement. In contrast, the ban on receipt of benefits for individuals with felony drug offenses is over-inclusive, because it disallows SNAP benefits to people who have never and would never engage in fraudulent use of SNAP or TANF benefits—for life.

Implications for Reentry and Recidivism

Each year, nearly 700,000 people are released from state and federal prison.³³ Along with the stigma of the criminal conviction and incarceration that they carry, a host of public policy restrictions make the reentry process increasingly challenging. In addition to potentially losing access to food stamps and TANF benefits, individuals with felony convictions (for drug offenses or other felonies, depending on the particular sanction) may not be eligible for public housing or federal loans to pursue an education; they may face substantial hurdles in obtaining employment, particularly when this involves applying for a professional license; driver’s licenses may be suspended; and there may be a loss of the right to vote, serve on a jury, or join the military.

These collateral consequences of a criminal conviction would be difficult to manage under any circumstances, but for people who are trying to reenter society after a period of incarceration, they are particularly damaging. Most people returning home from prison had been struggling in some significant way prior to their involvement with the criminal justice system; surveys consistently show that substantial proportions of people who are incarcerated have histories of substance abuse, mental health issues, homelessness, or physical or sexual abuse.³⁴ Without proper support, these individuals may continue to struggle with similar issues upon their release from prison.

Access to SNAP and TANF benefits may be particularly critical for individuals returning home from prison.

In this context, access to SNAP and TANF benefits may be particularly critical. The SNAP and TANF programs are designed to provide subsistence level benefits for people who cannot afford to feed themselves or clothe their children.³⁵ People who use these benefits typically do so for short periods of time; one overview of the program found that less than ten percent of recipients used food stamp benefits for five consecutive years.³⁶ People who apply for benefits are more likely to do so in the wake of a catastrophic life event, such as the loss of a job.³⁷ For formerly incarcerated individuals transitioning back to their home communities, SNAP or TANF benefits can help to meet their basic survival needs during the period in which they are searching for jobs or housing. By doing so, the programs reduce the likelihood that formerly incarcerated individuals will return to criminal activity to secure food or other essentials for themselves or their families.

Restrictions on SNAP and TANF benefits are also counterproductive for providing drug treatment services. Historically, drug treatment facilities have used their patients’ SNAP and TANF benefits to subsidize the cost of treatment.³⁸ If individuals who are recovering from drug addiction are denied access to these “subsistence benefits, treatment, and safe and sober housing, it is much less likely that these [people] will be able to live drug-free in the community and avoid recidivism.”³⁸

Public Health Effects

In addition to enhancing the risk of recidivism, there is some evidence that barring individuals with felony drug convictions from receiving food stamps may have troubling public health consequences. One of the few analyses done in this area was a recent pilot study conducted in Texas, California, and Connecticut that examined the relationship between “food insecurity and HIV risk behaviors among individuals recently released from U.S. prisons.”⁴⁰ The study found that formerly incarcerated people who lived in states that fully enforce the ban on receipt of food stamps for individuals with felony drug convictions were more likely to report having gone an entire day without eating than people who lived in states that did not enforce the ban; furthermore, people who did not eat for an entire day were more likely to engage in HIV risk behaviors, such as using alcohol, heroin, or cocaine before sex or exchanging sex for money.⁴¹ While the authors note that the small sample size limits the ability to draw definitive conclusions, they report that “[i]ndividuals released from prison are at high risk for food insecurity,” and that the level of food insecurity among recently released prisoners uncovered by the study “mirror[s] the magnitude of food insecurity in developing countries.”⁴²

Overall, there is little reason to believe that the drug felony ban has had any constructive impact on either substance abuse or public safety. States that enforce the ban in full have not conducted any studies that suggest there may be positive outcomes in comparison to states that have fully opted out of the ban. After 17 years of implementation, though, there is reason to believe that affected individuals in these states may be subject to substantial reentry challenges and food insecurity.

Current Political Climate

Since the TANF ban was enacted in 1996, a number of states have taken action to opt out of its provisions in full or in part, but three-quarters still retain either a full or partial ban on the receipt of welfare benefits. At the federal level members of Congress have introduced bills that would repeal the ban, but such legislation has not gained sufficient support to change policy.⁴³

More recently there have even been proposals to expand the scope of the ban’s restrictions, such as the one introduced during the 2013 legislative session of Congress by Senator David Vitter (R-LA). Senator Vitter’s proposal, which was presented as an amendment to the omnibus Farm Bill, called for a retroactive ban on individuals convicted of murder, aggravated sexual abuse, or sexual exploitation of children from receiving SNAP benefits for life.⁴⁴ Although the amendment was strongly denounced by many outside the halls of Congress,⁴⁵ it was unanimously consented to in the Senate.⁴⁶ A version of the amendment was later approved by the House as well.⁴⁷ However, broader political strife over the Farm Bill leaves the future of such policy uncertain.

Policy Recommendations

Since the adoption of the ban on receipt of SNAP and TANF benefits for individuals with felony drug convictions in 1996, we estimate that 180,100 women in the states that fully enforce the ban’s provisions may be affected by these provisions at some point in their lives. Including women in the states with partial bans, or men who are impacted by the policy, would clearly raise this number substantially.

There is no evidence to date that any harm caused by the ban has been offset by the realization of significant positive outcomes for public safety. The ban has not been shown to decrease drug use, nor is it necessary to reduce welfare fraud, which is proscribed by other sections of the United States Code. Furthermore, by raising a new substantial barrier to successful reentry, the ban may actually harm public safety and public health, while contributing to swollen prison populations. Policymakers who wish to address these challenges should consider the following reforms:

Congress

Given how little evidence was supplied in support of the ban in 1996 or regarding its impact since then, it is long overdue for Congress to repeal the drug felony ban on access to welfare benefits and food stamps. Among other incongruous effects, the ban is clearly inconsistent with Congressional support for reentry services through funding provided by the Second Chance Act, as well as current policy recommendations of the Federal Interagency Reentry Council. Policies such as the TANF/SNAP ban make it increasingly difficult for formerly incarcerated individuals to return home and lead productive law-abiding lives.

States

Until such time as Congressional repeal of the ban on receipt of SNAP and TANF benefits is enacted, states should consider adopting policies to opt out of the ban’s provisions. At a minimum, states should modify the ban such that individuals with felony drug convictions have some possibility of regaining eligibility for SNAP or TANF benefits—perhaps by successfully completing drug education or treatment. To the extent that any prohibitions remain in place, they should be narrowly tailored to achieving some kind of public health or safety goal, rather than being merely punitive in nature.