Report | Budget Taxes and Public Investment

When Work Just Isn’t Enough: Measuring hardships faced by families after moving from welfare to work

 Download PDF

Share this page:

When work just isn’t enough

Measuring hardships faced by families after moving from welfare to work

by Heather Boushey and Bethney Gundersen

Since the implementation of welfare reform via the 1996 Personal Responsibility and Work Opportunity Reconciliation Act, welfare caseloads have dropped dramatically, from 5.5% of the total U.S. population in 1994 to 2.1% in June 2000.1 The research on those who have left the welfare rolls has shown that a majority of former welfare recipients have worked at some point after leaving welfare, although most are earning only around $7.00 per hour (Assistant Secretary for Planning and Evaluation 2000; Brauner and Loprest 1999; Freedman et al. 2000; Loprest 1999; Parrott 1998). Recent data also show that, although poverty is down overall, it actually has deepened for those who remain poor and has increased among working families (Primus and Greenstein 2000).

But statistics on employment and poverty rates among welfare leavers may not provide a complete understanding of whether families are better off by moving from welfare to work. As families move off welfare, measuring the hardships they face is increasingly important to understanding whether these families are meeting their basic needs. Former welfare families who now work in the low-wage labor market are likely to have difficulty meeting work-associated expenses such as child care and transportation costs. Given the extra income often required just to participate in the workforce, these families also may have difficulty meeting other basic needs, such as food, housing, and medical care.

This report documents the extent to which families face hardships as they move from welfare to work. To this end, we compare how different types of families experience hardship, looking specifically at working-poor families, families that currently receive welfare, families that have recently left welfare, and families that left welfare over one year earlier.

Methods of measuring hardships

To measure the hardships that families experience while receiving welfare and after leaving the welfare rolls, it is necessary to understand how these families meet their basic needs. This is typically accomplished via surveys with questions about food sufficiency, evictions, and adequate heating in cold-weather months. Unfortunately, it may be difficult to ascertain whether the answers to these surveys indicate a genuine hardship or simply a family’s preference or taste. Therefore, the kinds of questions asked and the way in which they are worded becomes very important.

To help create accurate measures of hardship that reflect a family’s actual deprivation, research by Sondra Beverly proposes the use of material hardship variables (2000; 2001). According to Beverly, material hardships should reflect basic standards of material adequacy, assessing consumption of only food, housing, utilities, medical care, clothing, and consumer durables. She further recommends that the indicators should measure the severity of hardships in an objective framework. The core set of hardship measures should consist of direct, rather than indirect, indicators, and they should indicate the cause of hardship as well as include composite indices and separate measures of hardship. Such a method should weed out families that may “choose” to experience certain hardships. For example, not having a telephone is often considered a material hardship. However, some families may choose not to have a telephone for personal reasons. Alternatively, a wealthy person may not have paid the full amount of the rent or mortgage last month because he chose to spend the money on luxury goods instead. This person may not be experiencing a hardship in the same way as someone who did not pay the rent because he or she could not afford to. This, again, underscores the importance of question wording in the surveys.

The method used in this report expands on the notion of material hardships, which we refer to here as “critical” hardships, by looking as well at a broader array of what we call “serious” hardships. Critical hardships explain the extent to which families fail to meet their basic needs. In comparison, serious hardships explain the extent to which families are lacking in the goods, services, and financial ability to maintain employment and a stable, healthy home environment. The concept of serious hardships includes lack of access to regular and preventive medical care, low-quality and insufficient-quality child care, the inability to pay housing bills, and unreliable transportation.2

Hardship indicators

Hardships fall under four broad categories: food insecurity, insufficient access to health care, housing problems, and inadequate child care. To understand whether or not families are living with hardships, it is necessary to know whether basic needs under these categories are being met. This requires the use of surveys with detailed questions about how families meet their basic needs. Given the specific and invasive nature of the questions, the primary difficulty with measuring hardships is the availability and consistency of appropriate data.

There are two national surveys that ask questions about family hardships as well as welfare use – the Survey of Income and Program Participation (SIPP) and the National Survey of American Families (NSAF). These surveys ask families questions about whether they “go without” and experience material deprivation. This report uses both the SIPP and the NSAF in its analysis.3 The questions asked in the two surveys are not identical, but the use of both surveys allows us to expand our measures of hardship. Table 1 presents the questions we chose from the two datasets.

Table 1

While the SIPP and NSAF both have their strengths and weaknesses, we believe the nature and structure of the NSAF makes it a better survey from which to study hardships. The SIPP is a panel dataset focused on income, employment, and participation in government programs. The hardship questions are the last set of questions asked in the last month of the panel. As with all panel datasets, attrition is a problem in the SIPP. Since attrition tends to occur among low-income respondents, and these respondents are more likely to experience hardship, the SIPP measures may underestimate the extent of hardship. The NSAF, on the other hand, is a cross-sectional survey of the experiences of families that focuses on a number of hardships and family-stress indicators. The cross-sectional nature of this survey eliminates attrition bias. Additionally, the NSAF survey focuses exclusively on well-being, perhaps prompting respondents to give more thoughtful or detailed answers to hardship questions. Given the NSAF’s advantages in this regard, we focus on those results. The SIPP results are presented in the tables in order to demonstrate the similarity of the results across measures of hardship. (A complete analysis of these issues can be found in Gundersen and Boushey (2001) and Boushey et al. (2001).)

Critical hardships and serious hardships

Food insecurity
To go without sufficient food is the most basic critical hardship. We measure food insufficiency in terms of not having enough to eat or skipping meals sometimes or often. These measures are based on questions that have been tested for validity: families who report food insufficiency have been found to have lower food expenditures and lower intake of calories and nutrients (Rose 1999). A serious food hardship is less severe in that the family may not actually go without food, but it does not have the kind of food it would like or it worries about food intake.

The varia
bles that went into determining a critical food hardship include:

  • whether a respondent missed meals sometimes or often in the last 12 months (from the NSAF).
  • whether a respondent sometimes or often did not have enough food to eat (from the SIPP).

The variables that went into determining a serious food hardship include:

  • whether a family worries that food will run out before it can buy more (NSAF).
  • whether a family lacked the kinds of food it liked to eat (SIPP).

Housing problems
In accordance with Beverly’s work (2000; 2001), critical housing hardships are defined as the consequences of not paying housing bills. Eviction, utility disconnection, and moving in with others because a family could not pay its bills constitute critical housing hardships. Serious housing hardships, on the other hand, measure a family’s ability to afford housing or utility bills; these families may not have actually experienced a critical hardship such as an eviction in the past year, but they still struggle to make ends meet. Losing telephone service would be considered a serious hardship because a telephone is necessary to find and keep a job.

The variables that went into determining a critical hardship in housing include:

  • whether there was a time in the past 12 months that an individual was evicted from his or her household because the household did not pay the mortgage or rent (SIPP).
  • whether a family moved in with others because it could not afford to pay its mortgage, rent, or utility bill (NSAF).
  • whether there was a time in the past 12 months that an individual’s household had its utility service disconnected because the household did not pay the utility bill (SIPP).

The variables that went into determining a serious hardship in housing include:

  • whether there was a time in the past 12 months that an individual’s household was not able to pay the mortgage, rent, or utility bill (NSAF).
  • whether there was a time in the past 12 months that an individual’s household did not have telephone service because payments were not made to the telephone company (NSAF).

Insufficient access to health care
A critical hardship in health care occurs when, in the past 12 months, any individual in the family did not get or postponed necessary medical care. As Beverly (2000) notes, some consider this measure to be subjective, because respondents are asked to evaluate whether they “need” to see the doctor. However, we agree with others, such as Mayer and Jencks (1989), who argue that not being able to see a doctor when you feel you need to (whether you objectively need to see one or not) is a hardship.

While critical hardships involve immediate medical needs, the survey question gauging serious health care hardships inquires about access to preventative care. Having health insurance is a key element in one’s ability to access preventative care. However, even families with health insurance can still lack access to preventative care, either because their plans do not cover it or they cannot afford the out-of-pocket costs (Beverly 2000). To determine whether families are actually accessing preventative care, we consider whether or not families use an emergency room as their usual source of health care. Families who rely primarily on emergency room services for health care do not receive adequate preventative care, and the care they do receive is episodic, non-comprehensive, and lacks the benefit of a continuous relationship with a physician (Weissman and Epstein 1994). As a result, these families are classified as experiencing serious hardships.

A family experienced a critical hardship in health care if:

  • in the past 12 months any individual in the family did not get or postponed necessary medical care (NSAF).

A family experienced a serious health care hardship if:

  • any individual in the family used the emergency room as their usual place of health care (NSAF).

Inadequate child care
There are no critical hardships for child care because it is not considered an immediate basic need. Lack of adequate child care is, however, considered a serious hardship because it has long-term consequences for the family’s well being. A serious hardship in child care occurs if the parent experiences the hardship for any child in the household.

Measuring serious child care hardships involves an assessment of quality of care. Many researchers believe that child care quality has important effects on both the short- and long-term well-being of children (Vandell and Wolfe 2000). When using these measures, we assume that families are providing the best-quality child care they can afford.

We measure child care quality (reported in the NSAF) by the adult-to-child ratio in the child care setting, based on the recommendations of the American Academy of Pediatrics and the American Public Health Association.

For school-age children, we look at the type of non-school care a child receives, which has been shown to influence a variety of aspects of a child’s well-being (Capizzano, Adams, and Tout 2000). We use two measures of quality: whether the child cares for him or herself and whether the child is involved in activities. Children who care for themselves during non-school hours are placed at a greater risk for physical and psychological harm and are at a greater risk for being victims of crime. Self-care has also been linked to poor school performance, behavioral problems, and an increased chance of engaging in risky behaviors such as smoking, alcohol and drug use, sexual activity, and crime. Children involved in extracurricular activities and enrichment programs have been shown to perform better in school and to adjust better socially.

The variables used to determine if a family is experiencing a serious child care hardship include:

  • whether the adult-to-child ratio is less than that recommended by the American Academy of Pediatrics and the American Public Health Association (American Public Health Association and American Academy of Pediatrics 1992) (NSAF).
  • whether a child has cared for himself or herself in the past month or stayed alone with a sibling under 13 years old (NSAF).
  • whether a child was involved in any enrichment activities (NSAF).

In this report we examine only those families with positive income that are headed by an adult between the ages of 18 and 64.4

We divide families into four groups based on whether someone in the family currently receives welfare or has in the past.5 The first group consists of families who received Aid to Families With Dependent Children (AFDC) during the last month of either the SIPP or NSAF surveys. The SIPP and NSAF groups are comparable because both were on welfare prior to the implementation of the welfare reform act, although many recipients would be subjected to changes brought on by policy innovation in the states prior to the federal law.

The second group consists of families who received other public assistance (such as state General Assistance), but not AFDC, in the last month of either survey. The third family grouping, recent AFDC leavers, represents families that received AFDC anytime in the 12-month period prior to the survey. Finally, the fourth group includes those families who received AFDC at some point over a year ago.

Table 2 presents descriptive statistics for the SIPP and NSAF samples, including welfare status and poverty level.6 The samples are very similar across demographic groups. The table also presents unweighted sample sizes for the different categories, demonstrating that there are sufficient observations to analyze hardships across work status, public assistance status, and poverty status. Analyses not presented here
also demonstrated that there are sufficient observations to analyze hardship across work status for families with recent AFDC leavers and families that left AFDC more than one year ago, with the exception that there are not enough observations in the NSAF to analyze hardship for recent AFDC leavers who are not working.

Table 2

Do welfare and former welfare families experience more hardships than other families?

The prevalence of hardships
Before examining the hardships that welfare families experience, we look at the hardships experienced by a broader range of family types. (The appendix presents this information at the state level.) Table 3 shows the number of critical hardships and serious hardships families experience by poverty level. Around 30% of families living below the poverty level experienced one or more critical hardships, while only about 11% of families living above 200% of poverty experienced any critical hardships. The differences by income level are even more stark when looking at serious hardships. Among families living under the poverty line, about 77% experienced one or more serious hardships. However, among families living above 200% of the poverty level, only about 40% did.

Table 3

Table 4 shows the proportion of families who experienced critical or serious hardships by welfare use. A family currently on public assistance or receiving AFDC has more trouble meeting its basic needs than does the average poor family. More than 40% of families that currently receive public assistance (1.7 million people) experience at least one critical hardship. These families are even more likely to experience serious hardships, with over 85% (1.9 million people) experiencing at least one hardship of this type. AFDC families do a little bit better, with about 35% (4.3 million people) experiencing at least one critical hardship, and 78% (6.0 million people) experiencing at least one serious hardship.

Table 4

Families that have left AFDC more than a year ago have an easier time meeting their basic needs compared to current recipients or more recent AFDC leavers. Families who left AFDC more than one year ago are similar to poor families in their experience with critical hardship: less than 30% of each group experience at least one critical hardship. Recent AFDC leavers do slightly worse than poor families, with 35% experiencing at least one critical hardship, the same rate as current AFDC recipients. Recent leavers are similar to poor families in their experience with serious hardships: almost 80% of each group experienced at least one serious hardship. Families that left AFDC more than one year ago do better than poor families in terms of serious hardships, with only 70% experiencing at least one hardship of this kind.

The higher proportions of recent AFDC leavers who experienced hardships compared to those who left more than a year ago may be related to their work status. Table 5 shows the proportion of hardships experienced across the two groups of welfare leavers by work status.7 Families that left AFDC recently as well as those that left more than one year ago were less likely to experience hardship if they were working full time, full year. The two groups of families (both recent leavers and over one year ago) with less than full-time employment experienced similar levels of critical and serious hardships. This rate was higher than for poor families overall. Conversely, the families with at least one full-time, full-year worker experienced a lower rate of critical hardship than poor families. Comparing across the two groups of families who have left AFDC, families who left AFDC more than one year ago are somewhat better off than those who left within the past year, but only if they had a full-time worker in the family. Regardless of work status, however, these families continue to be no better off than other families living in poverty.

Table 5

The particular hardships that families experience
Table 6
shows the particular hardships experienced by families who received AFDC and public assistance in 1996. Public assistance families experienced far more hardships than other families. Strikingly, 65% of public assistance families (2.7 million people) worried about food, and 45% (nearly 1.9 million people) could not pay their rent, mortgage, or utilities. Among AFDC families, 24% had severe food hardships (3 million people), and 59% worried about food (7.5 million people). Over one-third of people were unable to pay their rent, mortgage, or utilities. Welfare families did not have considerably more trouble with child care hardships relative to poor families.

Table 6

Table 7 shows what kinds of critical and serious hardships poor families face based on whether there are any family members who perform paid employment. Surprisingly, critical hardships are just as prevalent among families who do some work (a family member working less than full time) relative to families who do not work or who have a family member who works full time. This may reflect the fact that poor families who do some work usually have no access to the government programs (or are unaware of their eligibility) that can help non-working poor families meet their needs. Poor families who work full time, on the other hand, may earn slightly more than the poor families who work less than full time, and the extra income enables them to meet their needs.

Table 7

Serious hardships among poor families of different work status do not exhibit as strong a trend. Child care problems are about the same regardless of work status for poor families, although non-working families are more likely to have their children in child care situations in which the adult-to-child ratio is below recommended standards. Having a non-working telephone is more common among poor families that do not work full time. This may be related to the work status of the family: families may have difficulty maintaining full-time employment without a working telephone. Worrying about food is also more common among poor families without a full-time worker. Poor families who do not work are less likely to have difficulty making their housing payments than are working poor families. This may reflect the fact that poor non-working families have more time to seek out rent and utility assistance than do working poor families.

Table 8 shows the particular hardships for families who recently left AFDC by work status, with trends similar to those found in Table 7. Of families who recently left AFDC, those with only part-time work experienced more critical and serious hardships than families with a full-time worker. The exception is that, families with a full-time worker are more likely to worry about food than families working part time. These results show that families who recently left AFDC still faced significant hardships in all areas, especially if they worked less than full time.

Table 8

Table 9 shows the particular hardships experienced by families who left AFDC over one year ago by work status. Families who were not working experienced more hardship than those who were. Three-quarters of non-working families worried abo
ut food, and nearly half were unable to pay their rent, mortgage, or utilities. One-fifth of non-working families had to double-up with friends or family. This is very high considering that only 3.4% of all families below 100% of poverty doubled-up in this way. Smaller proportions of families with a full-time worker experienced hardship relative to families working less than full time.

Table 9

Comparing families who recently left AFDC to those who left more than a year ago reveals that the more recent leavers experienced different kinds of hardships. Recent AFDC leavers who worked full time were more likely to worry about food than were comparable families that left AFDC over one year ago. However, the converse is true for those that worked part time. Recent leavers are more likely to experience housing problems than are families who left AFDC over one year ago, regardless of work status. Both recent leavers and those who left over one year ago experience more hardship across all kinds of work status than do families living below the poverty line.

These results are consistent with other research in this area. Pamela Loprest has compared welfare leavers in 1996 and 1998 using the NSAF data (Loprest 2001). She found relatively little difference among leavers across the two years. Some hardships among former welfare recipients were even more prevalent in 1998, relative to 1996. A survey of U.S. poverty conducted in early 2001 found that 42% of poor families fell behind on their rent or mortgage, 53% fell behind on their utility or phone bills, and 51% were unable to get necessary medical care (National Public Radio, Kaiser Family Foundation, and Kennedy School of Government 2001). These results show an even greater level of hardship than found in the NSAF for 1996, indicating that hardships among the poor actually may have increased over the 1990s, instead of declined, as would be expected during an economic boom.

There is little difference among families that left welfare recently when compared to those who left more than a year ago, unless the family contained a full-time worker. Even then, however, these families did not do as well as poor families more generally. Other findings include:

  • Poor families who work part-time have more difficulty, relative to poor families with a full-time worker or no worker at all, meeting needs related to housing, indicated by their higher rates of doubling up, inability to pay rent or utilities, and not having telephone service.
  • Families recently off AFDC who had a full-time worker were more likely than families with a part-time worker to worry about food and have their children in a child care facility with less than the recommended number of adults.

Much has been made of the success of welfare reform in the past couple of years. The dramatic declines in caseloads and the relatively high employment rates among families who have left welfare have led researchers to call the reforms a success. The reforms are only a success, however, if families that have left welfare are better off than they were on welfare. Our research indicates that, overall, former welfare families experience relatively high rates of hardship. Former welfare families with a full-time worker experience the lowest rate of hardship, but even among these “successful” families, work is not enough to ensure the family can meet its basic needs. Furthermore, not all former welfare families have a full-time worker, nor will all families be able to have one. In our analysis, 66% of families that left welfare more than a year ago and 57% of families that left welfare within the past year have someone in the family working full-time. As the economy moves into the next downswing, it may be increasingly difficult for these families to keep someone employed full time. But even families with a full-time worker experience more of certain kinds of hardships, such as worrying about food and inadequate adult supervision in their child care facilities. In short, families who moved from welfare to work still have trouble meeting their basic needs and therefore require work supports to help them avoid hardships.

This analysis makes it clear exactly what kinds of work supports such families need – food, housing, and medical care assistance. Although existing programs such as Food Stamps, housing subsidies, and Medicaid can assist former welfare recipients in avoiding hardship, there is certainly room to expand assistance in all of these areas.

Since 1996, the states have been allowed to expand Medicaid coverage via a waiver to working parents with incomes over the poverty line (Guyer and Mann 1999). But even so, only 11 states have expanded family coverage under Medicaid, and even then only to families earning less than 150% of the poverty line (Families USA 2000). Furthermore, many children eligible for state Children’s Health Insurance Program benefits are still not insured (Kenney, Ullman, and Weil 2000). States could make increased efforts to alleviate health care hardships through taking advantage of these programs.

As of 1998, up to one-third of welfare recipients received subsidized housing (Bryson 2000). Individuals who receive both welfare and housing assistance have been found to be more disadvantaged relative to those who receive only welfare (Kinglsey and Tatian 1999). Housing hardships could be reduced through increased efforts to integrate welfare-to-work services with housing providers.

This analysis also shows that these families have trouble meeting work-associated costs such as child care. The Child Care and Development Fund, which provides federal funding for child care assistance for low- and moderate-income families, has significantly increased the number of families receiving child care subsidies (Layzer and Collins 2000; U.S. Department of Human Services 1999). However, many families are still left without child care subsidies; in 1998, only 10-15% of federally eligible children received subsidies (U.S. Department of Human Services 1999). Several studies have documented long waiting lists for child care services and have found that many eligible families are not informed of these services by government agencies (Bell and Strege-Flora 2000; Human Services Alliance of Los Angeles 2000). States could decrease child care hardships for families as they move from welfare to work through increasing the coverage and public awareness of these child care subsidies.

Federal welfare reform law is scheduled to be considered for reauthorization in 2002. The results from this analysis should encourage policy makers to examine hardship measures, as well as income and employment, when trying to understand how former welfare recipients are faring. The analysis of hardships indicates that work alone is not enough to ensure families can meet their basic needs.

- June 2001

Tables A1.1 – A1.3 document the hardships experienced at the state level for 13 states. Sample sizes allow for an analysis of hardships only for families living in poverty (Table A1.1), families currently receiving AFDC (Table A1.2), and families who left welfare over one year ago (Table A1.3).

More poor families in Mississippi have trouble with serious (as opposed to critical) food and housing hardships than do families in the other analyzed states. Poor families in Michigan are more likely than those in other states to have critical hardship with respect to housing and are more likely to not have their children involved in activities. Poor families in Minnesota are less likely to experience food, health, and housing hardship compared to poor families in the other examined states. The overall trend, however, is that the experience of hardship across states for poor families is evenly distributed.

In terms of curre
nt and former welfare recipients, Texan families have more hardship than do families in other states. In particular, Texan families that received AFDC in 1996 are more likely to skip meals, not make housing payments, and have their telephone service disconnected. Families who receive welfare in Alabama are more likely to experience a health-related hardship compared to other states’ AFDC families. Overall, it appears that families who receive AFDC currently or in the past are having difficulty meeting basic and work-related needs, regardless of where they live.

Table A1.1

Table A1.2

Table A1.3

1. These data are available at

2. We do not have a measure for transportation hardship in our data. We could have included the lack of a vehicle, but this is an inadequate measure for two reasons. First, even families with vehicles could be driving an unsafe vehicle or lack car insurance. Second, in the few cities with adequate public transportation systems, lack of a vehicle is not a hardship.

3. The NSAF is a cross-sectional survey of the economic, health, and social characteristics of children, adults under the age of 65, and their families. It consists of data for 1997 and 1999. The sample is representative for 13 states (Alabama, California, Colorado, Florida, Massachusetts, Michigan, Minnesota, Mississippi, New Jersey, New York, Texas, Washington, and Wisconsin). At this time, we only have access to the 1997 survey data. For that year, NSAF respondents answered questions regarding hardships they experienced in 1996.

The SIPP is a multi-panel longitudinal survey of the civilian, non-institutional population in the United States. Respondents are interviewed every four months, and the data are compiled so that the respondents’ economic circumstances can be examined by month. The SIPP data for this study come from wave 9 of the 1993 panel. This wave refers to a four-month period in 1995. Information on income, demographics, and program participation come from the core wave and longitudinal files, while information on critical hardships and serious hardships come from the topical module associated with wave 9.

4. Less than 1% of SIPP households were headed by someone less than 18 years old, and they were dropped from the sample.

In the SIPP we use households as our unit of analysis because they are most similar in composition to the definition of social family that we use for the NSAF (Gundersen and Boushey 2001).

5. In our descriptive analysis, we use person weights. Thus, we examine the number of people who live in a family that experience critical and serious hardships.

6. The definition of poverty in the NSAF is whether the annual family income for 1996 falls above or below 200% of the poverty threshold. In the SIPP, the poverty threshold is a bit more complicated. The SIPP household poverty variable is an annual poverty level that changes each month because it is adjusted for inflation. This is summed over the past 12 months and divided by 12 to get the average poverty level over the course of the year. To determine if the household falls above or below 200% of poverty, we divide the household income variable by twice the average poverty threshold.

7. The sample size in the NSAF for recent AFDC leavers who do not work was too small to conduct reliable statistical analyses.

American Public Health Association, and American Academy of Pediatrics. 1992. National Health and Safety Performance Standards: Guidelines for Out-of-Home Child Care Programs. Arlington, Va.: National Center for Education in Maternal and Child Health.

Assistant Secretary for Planning and Evaluation. 2000. “Leavers” and Diversion Studies: Summary of Research on Welfare Outcomes Funded by ASPE. Washington, D.C. <>

Bell, Lissa, and Carson Strege-Flora. 2000. Access Denied. Washington, D.C.: National Campaign for Jobs and Income Support.

Beverly, Sondra. 2000. Using Measures of Material Hardship to Assess Well-Being. Focus. Vol. 21 No. 2, pp. 65-69.

Beverly, Sondra. 2001. Measures of Material Hardship: Rationale and Recommendations. Journal of Poverty. Vol. 5, No. 1. pp. 23-41.

Boushey, Heather, Bethney Gundersen, Chauna Brocht, and Jared Bernstein. 2001. Measuring Hardships. Washington, D.C: Economic Policy Institute. Forthcoming.

Brauner, Sarah, and Pamela Loprest. 1999. Where Are They Now? What States’ Studies of People Who Left Welfare Tell Us. Washington, D.C.: The Urban Institute.

Bryson, David B. 2000. Welfare and Housing: How Can the Housing Assistance Programs Help Welfare Recipients? Oakland, Calif.: National Housing Law Project.

Capizzano, Jeffrey, Gina Adams, and Kathryn Tout. 2000. Child Care Patterns of School-Age Children with Employed Mothers. Washington, D.C.: Urban Institute.

Families USA. 2000. Five Good Reasons for States to Expand Family Coverage. Washington, D.C.: Families USA.

Freedman, Stephen, Daniel Friedlander, Gayle Hamilton, JoAnn Rock, Marisa Mitchell, Jodi Nudelman, Amanda Schweder, and Laura Storto. 2000. Evaluating Alternative Welfare-to-Work Approaches: Two-Year Impacts for Eleven Programs. Washington, D.C.: U.S. Department of Health and Human Services and U.S. Department of Education.

Gundersen, Bethney, and Heather Boushey. 2001. A New Approach to Measuring Poverty: Testing the Reliability of Hardship Indicators. Washington, D.C.: Economic Policy Institute. Forthcoming.

Guyer, Jocelyn, and Cindy Mann. 1999. Employed But Not Insured. Washington, D.C.: Center on Budget and Policy Priorities.

Human Services Alliance of Los Angeles. 2000. Transforming Child Care from the Ground Up. Los Angeles, Calif.: Human Services Alliance of Los Angeles.

Kenney, Genevieve M., Frank C. Ullman, and Alan Weil. 2000. Three Years into SCHIP: What States Are and Are Not Spending. Washington, D.C.: The Urban Institute.

Kinglsey, G. Thomas, and Peter Tatian. 1999. “Housing and Welfare Reform: Geography Matters.” In S. Newman, ed., The Home Front: Implications of Welfare Reform for Housing Policy. Washington, D.C.: Urban Institute Press.

Layzer, Jean, and Ann Collins. 2000. National Study of Child Care for Low-Income Families, State and Community Substudy Interim Report. Washington, D.C: U.S. Department of Health and Human Services.

Loprest, Pamela. 1999. How Families That Left Welfare Are Doing: A National Picture. Washington, D.C.: Urban Institute. <>

Loprest, Pamela. 2001. How Are Families That Left Welfare Doing? A Comparison of Early and Recent Welfare Leavers. Series B, No. B-36. Washington, D.C.: The Urban Institute.

Mayer, S.E. and C. Jencks. 1989. Poverty and the Distribution of Material Hardship. The Journal of Human Resources. Vol. 24, pp. 99-113.

National Public Radio, Kaiser Family Foundation, and Kennedy School of Government. 2001. National Survey on Poverty in America.

Parrott, Sharon. 1998. Welfare Recipients Who Find Jobs: What Do We Know About Employment and Earnings? Washington, D.C.: Center on Budget and Policy Priorities.

Primus, Wendell, and Robert Greenstein. 2000. Analysis of Census Bureau’s Income and Poverty Report for 1999. Washington, D.C.: Center on Budget and Policy Priorities. <>

Rose, Donald. 1999. Economic determinants and dietary consequences of food insecurity and hunger in the U.S. Journal of Nutrition. Vol. 129, No. 2S, pp. 517S-20S.

U.S. Department of Human Services. 1999. National Study of Child Care for Low-Income Families. Washington, D.C.: U.S. Department of Human Services.

Vandell, Deborah Lowe, and Barbara Wolfe. 2000. Child Care Quality: Does It Matter and Does It Need to be Improved? Washington, DC: U.S. Department of Human Services.

Weissman, Joel, and Arnold Epstein. 1994. Falling Through the Safety Net. Baltimore: Johns Hopkins University Press.

See related work on Budget Taxes and Public Investment

See more work by Heather Boushey and Bethney Gundersen