Copyright 1996 Center on Budget and Policy Priorities. Preferred Citation: Center on Budget and Policy Priorities, "The Cato Institute Report On Welfare Benefits: Do Cato's California Numbers Add Up?" (Washington: March 7, 1996


March 7, 1996

A recent Cato Institute report concluded that in every state in the nation, welfare pays far more than a low-wage job. According to the report, "The value of the total package of benefits [received by AFDC recipients] relative to a job providing the same after-tax income ranges from a high of $36,400 in Hawaii to a low of $11,500 in Mississippi. In eight jurisdictions...welfare pays at least the equivalent of a $25,000 a year job."

The report estimated that in California, the typical AFDC family received benefits totaling $20,687 annually. Governor Pete Wilson has cited the report as justification for reductions he has proposed in welfare benefits.

Cato's conclusions are striking. They also are inaccurate -- the Cato report is replete with analytic errors. For the nation in general and for California in particular, the report paints a misleading picture both of the amount of benefits most AFDC families receive and of the supposed advantages from relying on welfare rather than working.

The Cato report substantially overstates the income that AFDC families receive. It counts benefits that only a small fraction of AFDC families get -- such as housing assistance, which fewer than one-tenth of AFDC families in California receive -- as though the typical AFDC family receives them. It incorrectly measures -- and overstates -- the value of other forms of assistance, including food stamps, low-income energy assistance, and WIC. It counts Medicaid as though it were income to a family, even though Medicaid is insurance that cannot be used to purchase food, clothing and shelter. Most analysts believe Medicaid should not be counted as income as Cato did.

When these problems with the Cato report are corrected, the typical family of three receiving AFDC and food stamps in California turns out to receive less than $10,000 in income, rather than $20,687 as Cato mistakenly alleges. This leaves such a family about $2,000 below the poverty line.

While overstating the income of AFDC families, the Cato report substantially understates the income that low-income working families receive, further skewing its findings. The report counts food stamps as income for welfare families but fails to count food stamps as income for low-income working families, who also are usually eligible for food stamp aid. The same error is committed for a range of programs available to both low-income working and welfare families; in virtually every case, Cato counts benefits from these programs as income only for welfare families.

Cato also counts Medicaid as income for welfare families while counting neither Medicaid nor employer-provided insurance as income for working families. Yet Census data for California show that 53 percent of children living in working poor families not receiving government cash assistance are covered by private health insurance or Medicaid.[1]

Thus, Cato counts housing assistance as income for California AFDC families when fewer than 10 percent of them receive a housing subsidy, but fails to count health insurance coverage as income for working poor families when half of the children in these families have coverage.

The Cato report concludes that families receiving AFDC in California are better off than families earning $11.59 an hour. In fact, a California family of three with a full-time worker earning $11.59 an hour typically had total income in 1994 that exceeded the income of a family on AFDC by more than $10,000. (The Cato report uses data for 1994; this analysis does the same.) Even a family of three with a full-time, minimum wage worker that receives no AFDC benefits has more income than a non-working family that does receive AFDC.

In short, Cato's conclusions that welfare pays much more than low-wage work -- and that AFDC families live far above the poverty line -- are inaccurate for California and virtually all other states. Based on the erroneous conclusion that welfare typically pays much more than low-wage work, however, the Cato report calls for a dramatic reduction in assistance for poor families.

California's Governor, Pete Wilson, has advanced a proposal consistent with Cato's recommendation, proposing substantial
across-the-board reductions in AFDC benefits. Governor Wilson's plan would cut benefits immediately, impose additional across-the-board reductions once a family has received aid for six months, and impose further cuts when the family has received assistance for 12 months. Assistance levels would be reduced for families in which parents are complying with work and job search requirements but are unable to find a job.

These benefit cuts would follow a series of other reductions in AFDC benefits in California since 1990. Benefits currently are 26 percent below their 1989-1990 level, after adjusting for inflation. If additional AFDC benefit cuts that have been enacted but are not yet in effect are implemented, the benefit reduction in California since 1989-1990 will equal 30 percent to 35 percent. If Governor Wilson's new proposals also become law, AFDC benefits will fall still further below the levels of just a few years ago.

In advancing his latest proposals in the welfare area, Governor Wilson has relied heavily on the Cato report. Materials the Wilson Administration distributed at the time of the Governor's State of the State address said, "According to a recent study by the Cato Institute, the total benefit package for an AFDC recipient is more than $20,000 a year." Several key state legislators have also recently cited the Cato study.

Policymakers and analysts from across the political spectrum agree that welfare reform is needed and that the welfare system should do more to reward work. The debate over how to transform AFDC into a work-based system which moves families from welfare to work, helps working poor families make ends meet, and provides a safety net for families in which parents try but are unable to find work is one of the most important policy debates facing the nation. As this debate move forward, it should be informed by accurate information, however, not misleading analysis.


The Cato report concludes that AFDC families in California typically receive benefits totaling $20,687. This substantially overstates the level of benefits available to most California families that receive AFDC.

The Cato report counts housing assistance, low-income energy assistance, and WIC benefits as benefits that California AFDC families typically receive. But AFDC families do not typically get all of these benefits. In California, 90 percent of AFDC families do not receive any form of housing assistance. Cato adds $6,413 to the income the typical California AFDC family receives on the erroneous assumption that the typical family get housing assistance.[2]

In addition, roughly 90 percent of children in California AFDC families do not receive WIC benefits. Yet Cato also adds WIC benefits into its calculation of the income a typical California AFDC family receives.

Reliable data do not exist to determine the proportion of AFDC families in California or other states that receive energy assistance through the Low Income Home Energy Assistance Program (LIHEAP). Some AFDC families receive this benefit; others do not. Cato assumes the typical AFDC family in every state receives these benefits as well.

The proportion of AFDC families that simultaneously receive housing assistance plus energy assistance plus WIC is tiny. In California, the proportion is likely to be something like one or two percent. Cato assumes, however, that the typical California AFDC family receives all three benefits. Furthermore, Cato assumes the typical California AFDC family receives WIC benefits for two children, even though nine of every ten children receiving AFDC in California receive no WIC benefits at all.

Cato's treatment of Medicaid creates other problems. The Cato report counts Medicaid benefits as income that AFDC families receive, showing Medicaid as increasing the income of California AFDC families by $2,784. While AFDC families do receive Medicaid, counting it as income is problematic. Medicaid is an insurance program; the payments it makes go to doctors and hospitals, not to the insured families. These Medicaid payments cannot be used to meet basic family expenses such as food, clothing, and shelter.

Earlier this year, a distinguished National Academy of Sciences panel issued a major report on how poverty should be measured. The panel concluded that the value of health insurance coverage, including both coverage through a government program and privately provided coverage, should not be considered income.[3]

Not only does Cato count certain benefits as income when the typical AFDC family does not receive them, but Cato also
exaggerates the average benefit levels these programs provide. The low-income energy assistance program is a case in point. According to a LIHEAP information memorandum the Department of Health and Human Services issued in March 1995 - the source Cato itself cites as the basis for its data on the energy assistance program -- the average benefit from the LIHEAP heating/cooling assistance program is $81 a year in California. Cato estimates the annual LIHEAP benefit that California AFDC families receive, however, as being $368.

The Cato authors arrive at this highly inflated number in part by assuming that the typical AFDC recipient in California receives not only heating/cooling assistance but also LIHEAP "crisis assistance." In fact, about 80 percent of the California households that receive energy assistance benefits get aid only for heating/cooling assistance and do not receive any crisis aid. Even if every household in California that received LIHEAP crisis assistance was an AFDC family, which is not the case, only 10 percent of all AFDC families would have received this aid. Cato counts it anyway as a benefit the typical AFDC family receives.

Cato's estimates of the food stamp benefits that AFDC families receive are overstated as well. A family's food stamp benefits depend in part on the amount the family pays for housing. A family that receives housing assistance gets substantially less in food stamps than a family that receives no rental subsidy. As noted, the prototype family that Cato uses in its study receives housing assistance. Such a family therefore would qualify for smaller food stamp benefits than the typical AFDC family does. But Cato assumes that the typical AFDC family simultaneously receives housing assistance and receives the food stamp benefit levels that
families get if they do not get housing assistance.

The typical California AFDC family of three living in subsidized housing receives food stamps totaling $152 per month. This is $62 less per month -- $744 less per year -- than the food stamp benefit that Cato assumes the typical AFDC family gets.

In short, the Cato report counts various benefits that at least 90 percent of AFDC families do not receive, improperly counts health insurance coverage as income, and substantially exaggerates the average benefit for many of the benefits it counts.


In addition to overstating the level of benefits that most AFDC families receive, the authors of the report ignore all means-tested benefits that low-income working families receive, with the sole exception of the Earned Income Tax Credit. By ignoring these benefits, Cato vastly overstates the wage level needed for work to be more remunerative than welfare.

While Cato counts Medicaid as though it increases the income of AFDC recipients, Cato does not consider the value of employer-provided health insurance for working families that are covered under an employer health plan. Cato counts cash plus Medicaid benefits when welfare families receive them, but counts wages only -- ignoring medical benefits -- when working families receive them.

Furthermore, Cato fails to acknowledge that many children in low-income working families that are not on AFDC receive Medicaid. Under federal law, all children under age six whose families have incomes below 133 percent of the poverty line are eligible for Medicaid. So are all children aged six through 11 whose families have incomes below 100 percent of the poverty line. Millions of children in low-income working families receive Medicaid.

Census data show that 53 percent of all children in California who live in working poor families that receive no government cash assistance are covered by either private health insurance or Medicaid. Cato ignores this fact, acting as though only AFDC families receive health care coverage.

Cato's treatment of the food stamp program also is problematic. Eligibility for the food stamp program is based on a household's income and asset levels. Working poor families without significant assets are eligible for food stamp benefits.

Moreover, unlike housing assistance or energy assistance, the food stamp program is an entitlement program; all who are eligible and apply for aid receive its benefits. In 1993, the food stamp program provided an average benefit of $181 per month -- or nearly $2,200 per year -- to some 134,000 working poor households with children in California.

Nevertheless, Cato fails to count food stamp benefits as income for working poor families. Food stamp benefits are counted as income only for AFDC families. (Please see the text box for a detailed discussion of how to account for food stamps.)

The Cato report also ignores the receipt of WIC benefits by many low-income working families. Eligibility for the WIC program is based on income and nutritional risk, not on AFDC receipt. In 1994, some 412,000 children receiving WIC in California -- or 70 percent of all children receiving WIC in the state -- had no AFDC income.[4] These are primarily children from low-income working families. Despite this, Cato counts WIC as income only for AFDC families and fails to count it as income for working families.

Cato's treatment of other benefits follows the same path. Some working families receive energy assistance and housing assistance. Cato declines to count these benefits as income to working families, counting the benefits as income just for AFDC families.

In their report, the Cato authors contend it is proper to include benefits such as housing assistance and WIC as income for AFDC families despite the fact that most AFDC recipients do not receive them. In defending this decision, the authors state, "We believe it was proper to include those benefits because at least some recipients in every state do receive them." It appears this same reasoning does not apply when considering whether to count benefits as income for working families.


The typical AFDC family receives AFDC and food stamps. Both these benefits should be counted when determining such a family's income. Other than Medicaid, these are the only significant benefits that a majority of AFDC families receive.

In 1994, a California family of three was eligible for a maximum AFDC benefit of $607 per month, or $7,284 per year. (Seventy percent of AFDC families in California include two or fewer children.)

A family of three receiving $607 in AFDC benefits would have received approximately $200 per month in food stamp benefits, or about $2,400 a year.[5]

The typical California AFDC family of three thus had cash and food stamp income of $9,684 in 1994. This placed the family more than $2,000 below the poverty line.[6]

When additional benefits other than housing are considered, a typical AFDC family that gets such benefits still would receive a total benefit package that leaves it well below the poverty line. For example, a family receiving WIC benefits for a child over age one or a pregnant woman would receive $400 to $500 a year in such benefits.[7] (Families receiving WIC benefits for an infant or for both an infant and a mother would receive higher WIC benefits. But families can receive these higher benefits only for a relatively short period of time. Moreover, the typical AFDC family does not include an infant.) As noted earlier, a family receiving LIHEAP benefits in California typically received $81 a year from that program in 1994, a figure that is likely to be smaller today since federal funding for LIHEAP has been cut. AFDC families of three that receive food stamps plus WIC plus energy assistance -- which most AFDC families do not get -- thus would typically receive less than $10,300 per year. This is still $1,500 below the poverty line and far below Cato's estimate that the typical California AFDC family receives $20,687.

Some 90 percent of California AFDC families do not receive any housing assistance. Even AFDC families that do receive housing assistance typically have incomes well below Cato's $20,687 figure. Housing assistance saves California AFDC families that receive such aid an average of approximately $2,300 a year by reducing the out-of-pocket expenses these families incur for housing. (In calculating the value of housing subsidies, we have followed the Census Bureau's approach of counting the amount that a family saves on housing costs due to receipt of a housing subsidy as the amount by which the subsidy boosts the family's

A family receiving housing assistance is unlikely also to receive WIC and LIHEAP benefits. The 10 percent of California AFDC families that receive housing assistance thus typically have total income a few hundred dollars above the federal poverty line. It also should be noted that the poverty line is uniform nationally and does not reflect higher-than-average living costs in California. Were the poverty line adjusted to reflect variations in living costs, California AFDC families that receive housing assistance would likely be below it.

(Note: the cost to the government of providing subsidized housing is often higher than the amount a family saves on housing costs. Some families that receive rental certificates and vouchers to help pay the rent on private apartments may be able to afford somewhat higher quality housing as a result, such as housing that is less crowded or in a less dangerous neighborhood. In these cases, the value to the family may be greater than the amount the family saves by living in subsidized housing. On the other hand, for a substantial number of families living in public housing, the value of the subsidy to the family may be less than the amount the family saves on rent. Some public housing may be of poorer quality or in a more dangerous area than the private housing that families would rent if they paid more and did not live in public housing. If the housing is of lower quality than the housing a family would have rented without the subsidy, the amount the family saves may overstate the value the family receives from the subsidy.)


The Cato report suggests that a family receiving AFDC would not be made better off if the parent were to find a low-wage job. While most observers agree that the AFDC system does too little to encourage and reward work, the Cato report is inaccurate on this point. A family of three with a full-time, minimum wage worker generally has income higher than a non-working family receiving AFDC despite Cato's claim that an AFDC recipient would need to find a job paying $11.59 per hour to be "better-off."

A family with a full-time, year-round minimum wage worker has gross earnings of $8,840. (The federal minimum wage is $4.25 per hour.) Such a family pays FICA taxes (that is, Social Security and Medicare payroll taxes) but also receives the Earned Income Tax Credit.

In 1994, a family of three consisting of two children and a parent working full time throughout the year at the minimum wage received a $2,528 EITC payment and paid $676 in FICA tax, for a net gain of approximately $1,850. The family also would qualifyfor food stamps unless its assets were too high.

When food stamps, the earned income credit, and FICA taxes are considered, a family of three with a full-time minimum wage worker and limited assets could have received total income of about $13,400 in 1994. This exceeds the income of an AFDC family of three in California by more than $3,000.[10]

Furthermore, as part of the 1993 budget law, the Earned Income Tax Credit is expanding to make work still more remunerative than welfare. In 1996, the year in which the 1993 EITC expansions are phased in fully, a family with two children in which the parent works full time at the minimum wage will qualify for an EITC of $3,536. This expansion adds another $1,000 to the income of the minimum wage family, causing its total income to exceed that of the non-working AFDC family by more than $4,000.

As noted, large numbers of working poor and near-poor families also are eligible for benefits such as Medicaid, employer-paid health coverage, WIC, energy assistance and housing assistance. While working poor families are less likely to receive these benefits (except for employer-paid health insurance) than non-working families are -- and while the typical working poor family, like the typical non-working poor family, is very unlikely to benefit from all these programs -- these benefits do help substantial numbers of poor and near-poor working families meet basic needs.

Although a family of three with a full-time minimum wage worker receiving no AFDC income would generally have income higher than a non-working family receiving AFDC, some families with a full-time minimum wage worker would still be eligible for a modest income supplement from the AFDC program. California permits AFDC families that find jobs to keep more of their earnings than do many other states. If an AFDC recipient found a full-time, minimum wage job in 1994, the family would have been eligible for $300 per month in AFDC benefits (offset in part by reduced food stamp benefits).

In this way, the California AFDC program does a somewhat better job than the programs in other states of rewarding work.[11] Indeed, the structure of California's AFDC program provides parents with a significant financial incentive to find work even if they can find only part-time, low-wage jobs.

Working families that remain eligible for AFDC also receive assistance in paying for child care expenses, and they continue to qualify for Medicaid coverage. Families that find jobs in which they earn too much to be eligible for AFDC are eligible for a year of transitional Medicaid coverage and transitional child care assistance.

This is not to suggest that the safety net for working poor families is as strong as it should be. The programs' asset limits frequently prevent the working poor from qualifying for food stamps or AFDC (these limits are more stringent for AFDC eligibility than for food stamps), and working poor families that have not recently left welfare for work do not benefit from transitional Medicaid or child care assistance. In addition, while virtually all poor children born after September 1983 are eligible for Medicaid, including children in working poor families not on AFDC, the parents in these working families often lack health insurance for themselves. Most states, including California, also lack sufficient child care resources to provide child care subsidies to most low-income working families needing such assistance.

Moreover, the value of the minimum wage in 1996 will be 28 percent below its average value in the 1970s, after adjusting for inflation. This year or next, the purchasing power of the minimum wage will fall to its lowest level since 1955.

Policies to assist the working poor need strengthening. But the problems facing the working poor should not lead to the erroneous conclusion that non-working families receiving AFDC have higher incomes than families with a low-wage worker.


The authors of the Cato report make serious methodological errors in their analysis of the benefits that both AFDC families and working families receive. As a result, their estimates of the amount of earnings that working families need to be better off than welfare families are highly exaggerated. The authors inflate the value of basic assistance available to most families that receive AFDC while ignoring various forms of assistance available to low-income working families.

These mistakes lead the authors to the erroneous conclusion that families receiving AFDC are better off than families with substantial earnings and to a policy recommendation that benefits for poor families receiving cash assistance should be reduced dramatically. The authors fail to note that this type of "reform" has essentially been tried across the nation over the past two decades -- a period during which the combined value of AFDC and food stamps in the typical state has fallen more than one-quarter in purchasing power -- without producing the desired results in reforming welfare. The principal impact of this reduction in benefits has been to make poor families with children poorer.

In addition, while the report correctly points out that many families with a low-wage worker remain poor and face difficult challenges, it offers no suggestions for assisting these working families. It provides no recommendations to enhance the advantages of work over welfare in ways that help struggling working families to make ends meet.

As debate over welfare reform progresses at federal and state levels, participants in the debate need to know that the data they use and the claims they make are accurate. Policymakers and analysts also need to search for ways to assist working families trying to raise children on low wages. The Cato report provides no help on either front.


1. Government cash assistance includes both Supplemental Security Income for poor elderly and disabled people and AFDC. Recipients of both of these programs are automatically eligible for Medicaid under current law.

2. Cato acknowledges that most AFDC families do not receive a housing subsidy on page 28 of its report but otherwise ignores this fact.

3. The Census Bureau publishes a number of alternate measures of poverty. A few of the measures count Medicaid as income. Compared to these measure as well, the Cato method greatly inflates the amount by which Medicaid raises the incomes of California AFDC families.

Under these alternative Census Bureau poverty measures, the Census Bureau counts as income the amount of resources that
Medicaid frees up for other uses. Under this approach, the Census Bureau considers Medicaid to increase family income if the family's other income is sufficient to meet the family's basic housing and food needs. The Census Bureau compares a family's income to the Department of Agriculture's Thrifty Food Plan and the Department of Housing and Urban Development's measure of the "fair market rent" to determine whether the family's income is adequate to meet these needs. In 1994, the combined value of the Thrifty Food Plan and the fair market rent for a family of three in California was slightly above $14,000. Since the typical AFDC family of three in California has income below this threshold, there is not more income than needed to meet these basic needs. As a result, when using its alternative measures of poverty that include non-cash benefits, the Census Bureau does not assign an income value to receipt of Medicaid by AFDC families in California.

4. These figures include both infants and children under age five receiving WIC.

5. The food stamp benefit for a California family of three with AFDC income of $607 per month was calculated using the median shelter cost for three-person California families that received all of their cash income from AFDC and were not among the 10 percent of families with the lowest shelter costs. The 10 percent of families with the lowest shelter costs was excluded to screen out most AFDC families that receive housing assistance. Had these families not been screened out, the average food stamp benefit for AFDC families would have been lower.

In 1993, the median housing cost of those three-person in California families that received all of their cash income from AFDC and were not among the 10 percent of families with the lowest shelter costs was $385 per month. This figure was adjusted for inflation between 1993 and 1994 to calculate the food stamp benefits an AFDC family would have received in 1994.

6. In 1994, the poverty line for a family of three was $11,821.

7. In 1994, the average monthly value of the WIC food package in California was $34 for children, $43 for pregnant and post-partum women, and $70 for infants. An AFDC family receiving benefits for a child would receive average benefits of $411 a year. A family receiving benefits for an infant would receive about $843 a year. Data are from "Study of WIC Participant and Program Characteristics, 1994," December 1995, Abt Associates.

8. The methodology used here to determine the value of housing assistance is similar to that used by the Census Bureau when it calculates the value of housing assistance for families. Each year, the Census Bureau releases poverty data using alternative measures of income. Some of these measures include the value of housing assistance. In these measures, the Census Bureau calculates the value of housing assistance for a family by taking the difference between the housing cost the family faces with its subsidy and the estimated cost of housing it would face if it did not receive a housing subsidy.

9. A family of three in California with AFDC income of $607 per month would pay approximately $158 per month for housing if the family received a federal rent subsidy or lived in public housing. This is determined by applying the federal rules regarding tenant rent contributions in public housing and in the Section 8 rental assistance program. The median housing cost for a family of three that receives all of its cash income from AFDC was $385 per month in 1993 when the 10 percent of families with the lowest housing costs are excluded. Adjusted for inflation, this figure is $397 in 1994 dollars.

At first glance, housing assistance thus appears to increase the family's monthly income by $239, since the $397 median monthly housing cost the family would otherwise face is reduced by $239 by receipt of housing assistance. However, food stamp benefits are partially determined by a family's housing costs, and the food stamp benefits an AFDC family receives are lower if the family gets housing assistance. A typical California AFDC family of three that receives housing assistance would receive food stamp benefits $47 a month smaller than the food stamp benefits paid to a similar family that does not receive housing assistance and spends $397 per month on housing. Thus, receipt of housing assistance increases a typical AFDC family's disposable income by a net total of approximately $192 per month, or about $2,300 per year.

10. Similar to the methodology used in calculating the food stamp benefits for AFDC families, the level of food stamp benefits for this family was calculated using the median housing cost in California for families of three with earnings that receive food stamps.

11. While a family that receives AFDC and then finds a full-time, minimum wage job would be eligible for AFDC, families that were not first on AFDC may not be. The AFDC program in California treats earnings differently if the family found the job prior to applying for aid.

Copyright 1996 by Center on Budget and policy Priorities. Readers may redistribute this article to other individuals for noncommercial use, provided that the text and this notice remain intact. This article may not be resold, reprinted, or redistributed for compensation of any kind without prior written permission from the author. If you have any questions about permissions, please contact The Electronic Policy Network, P.O. Box 383080, Cambridge, MA 02238,, or by phone at (617) 547- 2950.

Return to Main Text