Effective Work Programs

Testimony Welfare

Effective Work Programs

April 9, 2002 14 min read
Jason Turner
Visiting Fellow in Welfare Policy
Visiting Fellow in Welfare Policy

Greetings to Members of the Committee:

Thank you for this opportunity to testify, specifically on the subjects of the TANF work requirements and the child care provisions. Let me first state that my testimony before this committee represents my personal views on this issue, and should not be construed as representing any official position of The Heritage Foundation.

As deliberations begin, it is worth reflecting on the enormous contribution the Congress has made to American families and our society through its enactment of the PRWORA reform which created the TANF program. The Heritage Foundation has compiled figures delineating some of the most significant achievements which have resulted:

Poverty has been reduced dramatically over a short period of time

Since the enactment of welfare reform in 1996, the conventional poverty rate fell from 13.7% in 1996 to 11.8% in 1999. There are 4.2 million fewer people living in poverty today than there were when the welfare reform law was enacted. Moreover when the EITC and non-cash welfare benefits such as Food Stamps and public housing are counted in determining poverty, the poverty rate in 1999 was even lower, at 8.8%.

Poverty among single mothers is at its lowest point in U.S. history. The rate has fallen from 44% in the mid-1990's to 35.7% in 1999, and to 25.7% when the EITC and non-cash benefits are included.

Child poverty has fallen from 20.5% in 1996 to 16.9% in 1999, and to 11.2% when the EITC and non-cash welfare benefits are counted as income. Child poverty declines have been greatest among African-American children. The conventional black child poverty rate has fallen by one-third, from around 43.8% in the mid-1990's to 33.1% in 1999, and when the EITC and non-cash benefits are included it is 21.6%. And the number of children living in "deep poverty" (incomes less than half the poverty level) declined from 6.3 million children in 1996 to 4.9 million just three years later in 1999.

Employment among single mothers is way up

The employment effects of PWRORA have been felt most strongly among the least employment-ready single parents. Since the mid-1990's the employment of never-married mothers has increased almost fifty percent, that of single mothers who are high school dropouts by two-thirds, and that of young single mothers aged 18-24 by a doubling.

Welfare dependency has plummeted

The national welfare caseload has declined by over 56%, dropping from 4.4 million cases in August of 1996 to 2.1 million in September of 2001. Moreover dependence has fallen most sharply among young never-married mothers with young children who have low levels of education, demonstrating that the reforms have affected the entire caseload, not merely the most employable.

There has been a halt in the rise of out-of-wedlock childbearing and an early and small shift toward marriage

Overall out-of-wedlock births rose from 7.7% of all births in 1965 to an astonishing 32.6% in 1994. Thankfully, the relentless 30 year rise has at last taken a pause, and among African-Americans the out-of-wedlock birthrate has actually fallen slightly from 70.4% in 1994 to 68.8% in 1999. In addition the share of children living with married couples among blacks rose from 34.8% to 38.9%. While these changes are small overall, they represent the first hope that the decline in family stability experienced since the 1960's may be subject to reversal.


In the discussion below I will make two arguments as follows:

  • The reauthorized bill should include strengthened work requirements. These requirements are essential to transforming the meaning of welfare away from a cash entitlement, and to maximizing the rate of movement into and up within the private labor force. The work requirement rates in current law are obsolete and have been overtaken by events.
  • The budget for the reauthorized TANF program can be reduced by ten percent without adversely affecting any essential aspect of the program, including the provision of child care for working families, and would in many respects result in improvements in the effectiveness of the service delivery system.


The TANF program has been extraordinarily successful at reducing the caseload and moving individuals into employment, as we have seen above. State programs have achieved this by instituting good up-front job search programs in what is termed as a "Work First" approach. Experimental research over the past decade and a half, influential among the drafters of the current law, had revealed that education and training alone is less effective at helping individuals succeed in the private labor market than early entry into employment if feasible, where on-the-job learning can help individuals move up the employment ladder faster than holding them out of the labor market for classroom instruction. Most often actual work can be combined with education and training in a more effective combination than either one alone.

From this "Work First" orientation, our experience has shown further that for those unable to find immediate private employment, either full or part time, the next best alternative usually includes some work experience as a core part, although not the only part, of an overall schedule and effort resulting in employment. This is especially true for those without extensive prior work history.

There are two key components which together influence the effectiveness of welfare-to-work programs under TANF. One component is the number of hours of activity required of a participant, which is a measure of his or her effort. The second is the overall proportion of individuals engaged in such activities, which is a measure of the breadth and reach of the program. Both components, the intensity and the breadth of program participation, are important to the overall effectiveness of the program. The authors of the current TANF program clearly intended that both program intensity and program breadth be the focus, and they did so by setting meaningful levels of weekly work requirements (measured in hours), and participation rates (measured by the proportion of adults actually engaged in the activity).

Surprisingly, given the goals of TANF, the proportion of adults engaged in constructive activities leading to employment, is quite low, once those who are already employed while on welfare are excluded. Although over 40 percent of the adult caseload in the average state is involved in some required activity, nearly 70 percent of these are in unsubsidized employment; i.e., they are collecting welfare while working at a regular job. This is, of course, good as far as it goes. But for the remainder, i.e. those not working and still receiving benefits, current law has done little to encourage states to constructively engage this group.

In order for the TANF program to make significant continued progress at helping adult recipients achieve financial independence, it will have to find ways to get states to engage a far larger proportion of the adult population than is being served under the current program. A major management commitment is necessary to mount a large and ongoing program for a high proportion of recipients, and although the policy makers who drafted the TANF program may have anticipated that most recipients would be involved in welfare to work activities, implementation by states has simply not produced this result.

The President's TANF reauthorization proposal, Working Toward Independence, moves us in the right direction toward the next level of reform by focusing state programs on increasing the level of effort made by individuals in the program, and by increasing the program's breadth and reach. It does this while retaining the state operational flexibility inherent in the TANF program.

The President's plan sets a 40 hour week as the standard for welfare-to-work activity, which is an increase from 30 hours per week under current law (or 20 for single parents of children under 6). The 40 hour week is comparable to the time commitment necessary in a full-time job. Unlike current law, however, which measures only participant time spent in work-like activities such as subsidized employment and work experience, the President's plan divides required activity into two parts - - work-like activity for 24 hours per week (i.e. three-day equivalent) and state-flexible activity for the other 16 hours. This is intended to give states the flexibility they need to blend other program components into the week to maximize its effectiveness, such as education, training, substance abuse treatment, and job search.

In addition to moving to a higher level of participant weekly commitment, the President's plan also intends to increase the proportion of individuals actually engaged in welfare-to-work activities by gradually increasing the state required participation rate to 70%, from its current 50%, while making certain adjustments to the rate calculation to make it easier for states to achieve.

However, in our judgment the President's plan is in error by permitting states to count as participating those who have exited the program by reason of employment for up to three months. Although it appears to reward success, this provision inadvertently credits welfare programs with high benefit levels over those with lower benefits. Programs with high benefit levels tend to keep individuals in the welfare system while they are working (especially if they also have high income disregards). A better approach is to significantly increase the caseload reduction credit, thereby rewarding states which are able to find employment for their participants and reduce dependency at the same time.

Are the state work requirements as outlined in the President's plan realistic and achievable for the majority of states? Absolutely!

States have already shown from the current legislation that they are capable of designing programs to meet federal performance targets when challenged to do so. The President's plan sets important targets, but leaves the bulk of the operational decision-making to state policy makers.

Both former Governor Thompson of Wisconsin and Mayor Giuliani of New York City have designed and operated large-scale welfare-to-work programs as originally envisioned by the authors of PWRORA, and as likely to be achieved in practice under the President's bill (with certain suggested modifications). Both Wisconsin and New York share the aspiration to run full-week programs with high levels of required participation. Some of the practical fundamentals of operating such programs are outlined below:

Welfare-to-work programs should constitute genuine practice for private employment.

  • The program should operate on a standard full-time workweek which conforms to the expectations of private employment. This allows participants to practice
    organizing their lives around a realistic work schedule of eight hour work days and five day work weeks;
  • Real work should be made part of the weekly activity. The pride and satisfaction of successfully mastering work tasks often results in a big psychological lift and translates into confidence in the search for private employment;
  • Work assignments must include close supervision and regular feedback. Those who lack work histories are often not familiar with workplace norms of professionalism and conduct, and frequently find it difficult to submit to
    supervisory authority or get along with co-workers. Good supervisors who agree to make part of their task the acculturation of participants play a large role in the
    success of their charges;
  • There must be swift consequences for non-attendance without cause. The notion of such consequences can be a new and ultimately constructive experience for those used to being involved in a bureaucratic welfare system in which not much changes. Thus, the importance of reliability must be taught, and for this to occur benefits must be closely tied to attendance.

High levels of required and ongoing participation best allows for the goal of replacing cash assistance with work. Welfare-to-work activities which become part of an ongoing obligation as a condition of receipt of welfare, allow for an ever-present option for those rotating in and out of the labor market. It can operate much like an accordion, expanding and contracting to accommodate those out of the labor force, while keeping work habits and skills in good repair.

Required ongoing participant activity probably exerts its greatest net caseload impact at the time of enrollment. Where participation in welfare-to-work programs has been required of applicants who do not find private employment within a certain period of time, the number of actual slots used by participants is almost always far fewer than anticipated. Fewer slots are necessary because individuals who know they must engage in work in exchange for benefits frequently elect not to enroll in the program in the first place. Instead, they find immediate employment or increase their hours in existing part-time employment.

Universal work programs require work slots for individuals of all capabilities. In New York City, after excluding child-only cases in which there is no adult, fully 87 percent of the TANF caseload is deemed "engageable," meaning that they are ready for some kind of welfare-to-work assignment. Having a near-universal  expectation of work helps change the culture of the system and channels the energy of recipients in a constructive direction away from attempting to qualify for exemptions.

Sanction policies play a large role in achieving high levels of participation. High non-participation rates are a feature of most mandatory programs. In Wisconsin, where the W-2 program pays cash benefits only to those who first participate in work activities, compliance by definition is high. However, in states like New York that do not use a version of full-check sanction for non-participation, a large proportion of families may accept a lower TANF payment rather than engage in work.

High turnover rates present management problems but lower the number of required work slots. The high turnover rate has at least two causes. One cause is that those who reliably participate in their work assignments, even for short periods, find they can obtain private employment. Fully half of all individuals who participated in New York's work experience program for any period during the first quarter of 2000 found employment the same calendar year. In addition, normal caseload dynamics in which recipients leave the rolls further increases turnover. The high work experience turnover rate means that far fewer actual slots are needed to run a universal program than would otherwise be required.

In conclusion, managing a large-scale welfare-to-work program is both practical and necessary to achieving true welfare reform. The President's plan, with modifications, sets us in the right direction.


There is far more money available for welfare-to-work expenditures than ever before because about half of the prior expenditures on benefits are no longer required as a result of caseload reductions. This of course is a good development overall, and accommodates increased spending per remaining adult recipient, as well as permitting more funds to be dedicated to child care for working families, and other such supports.

However, we may be reaching a point where the plentiful availability of resources may begin to be counterproductive. The excess liquidity in the TANF system can result in programs being less efficient and effective than they otherwise might be if careful use of resources remains a budget necessity. For example, in New York City we now spend about ten times the amount per remaining recipient on welfare-to-work services (of all kinds, including child care and substance abuse treatment) as compared to prior to the passage of TANF, even though caseloads are about 60% lower (not ten times lower). This anomaly occurs because benefit payments represented the overwhelming proportion of total welfare spending in the pre-TANF era.

The significant increase in available funds has resulted in enormous pressure for states to find ways to spend or obligate funds. In a ten-state study published about six months ago, the General Accounting Office found that of ten states studied, five had used between fifteen and twenty five percent of their TANF funds to supplant state spending. Moreover, even with the pressure to expend funds, as recently as the first half of fiscal 2001 states as a group were spending at a rate equal to only 91% of their available block and supplemental TANF grants (states have now caught up and are spending at a rate slightly higher than that available through annual grants).

Another way to see the increase in available resources as a result of the caseload decline is to consider that from FY 1998 to FY2001, spending on cash assistance declined from 61% of total TANF expenditures to 38%. As a result, significant amounts of funds have been freed up for other uses. However, even counting all the 2001 spending on cash assistance - - for child care for welfare and working families, for education and training, for state supplements to the EITC, and for all other direct work supports - - there still remained 23% of the TANF block grant which was available and re-programmed for other uses, according to calculations made by the Center for Budget and Policy Priorities.

The result of excess liquidity in the TANF program means, for a state and local administrator, pressure to spend money in ways they might not otherwise deem wise. Some state and local administrators have had difficulty extracting the best value from employment and training vendors.

As of the end of the last fiscal year 7.4 billion dollars in federal funds remained as unobligated or unliquidated from the TANF block grant, or an accumulation rate of about 1.5 billion per year (unliquidated funds be committed, see footnote). A ten percent reduction would take out about 1.7 billion dollars per year the amount of federal funds otherwise available, or an amount not much greater than the excess which has accumulated each year.

Nor is there a shortage of child care funding. For FY 2002 the total federal share of child care funds through the CCDF, TANF and SSBG equals a very generous $8.7 billion. To this add the state shares under TANF and CCDF for a combined total of $11.7 billion. This amount does not account for children being cared for while participating in Head Start (another $6.5 billion) .

But even these figures underestimate the amount of federal resources devoted to supporting children in care arrangements. The dependent care tax credit subsidizes child care in an amount in excess of $2.6 billion (1998) per year. Moreover the two largest tax programs which help support children, the Earned Income Tax Credit and the Child Tax Credit, dwarf all other programs combined. The refundable EITC, originally conceived as one way to help low-income working families better manage the expenses of working (including the expense of child care), contributes over $50 billion to families per year. Finally, the child tax credit contributes over $20 billion to families.

The two systems, the direct subsidy system and the tax system, work together, with welfare parents and entry level employed adults relying more on direct subsidies, and low and middle income working families utilizing the tax subsidies to a greater extent.

Thirty-two states have no waiting lists for CCDF child care. Of those remaining that do, these states tend to have state criteria which extends eligibility way up into the middle class (e.g. California with a maximum income limit of $35,100, New Jersey at $36,570 and No. Carolina at $34,224). For those well into the middle class, states may wish to assure parents are utilizing the tax subsidy system while reserving its direct subsidies for its lower income families.

Finally, experience shows that child care waiting lists, particularly in large cities, are not always accurate. Maintaining lists is often complicated and bureaucratic. When New York City carefully went through its extensive waiting list, it found far fewer families actually needing child care than was implied by the size of the list. Reasons for this included the following:

  • Many families receiving child care subsidies no longer needed them because the child was no longer living with the family.
  • Some previously eligible for care for reasons of work or program participation were no longer engaged in the activity which provided their eligibility.
  • Some families were receiving one kind of child care subsidy, but were looking for another kind of care, e.g. a particular center.
  • Some families had placed their names several times on one or more lists.

In conclusion, the tremendous success of PWRORA at helping families achieve self-sufficiency has reduced the level of state and local funds necessary to provide benefit payments. The federal taxpayer should participate in at least some of this success in the form of reduced contributions to the TANF block grant.

The Heritage Foundation is a public policy, research, and educational organization recognized as exempt under section 501(c)(3) of the Internal Revenue Code. It is privately supported and receives no funds from any government at any level, nor does it perform any government or other contract work.

The Heritage Foundation is the most broadly supported think tank in the United States. During 2013, it had nearly 600,000 individual, foundation, and corporate supporters representing every state in the U.S. Its 2013 income came from the following sources:

Individuals 80%

Foundations 17%

Corporations 3%

The top five corporate givers provided The Heritage Foundation with 2% of its 2013 income. The Heritage Foundation’s books are audited annually by the national accounting firm of McGladrey, LLP.

Members of The Heritage Foundation staff testify as individuals discussing their own independent research. The views expressed are their own and do not reflect an institutional position for The Heritage Foundation or its board of trustees.


Jason Turner

Visiting Fellow in Welfare Policy