Few elected officials are eager to defend the U.S. social welfare system, and many Americans assume our safety net programs are a failure. The passage of the new welfare law will transform the shape of the safety net, giving state policymakers unprecedented authority to redesign low-income programs. Given the large consequences of these changes, it is important for the public to understand how the federal low-income assistance system has performed in recent decades.

Using analysis of recently released Census Bureau data, this paper shows that federal and state anti-poverty programs have lifted millions of children and disabled and elderly people out of poverty. Many of those who remained poor were significantly less poor than they would have been without government assistance.

The Census data show that both the number and the percentage of individuals removed from poverty by safety net programs either set or tied an all-time high in 1995. Without government programs, 57.6 million people would have been poor last year. But when government benefits are counted — including food stamps, housing assistance, school lunch support, and benefits provided through the earned income tax credit — the number of poor people drops to 30.3 million. In other words, the safety net programs lifted 27 million people out of poverty last year, cutting the size of the poverty population nearly 50 percent.

Using the definition of poverty in this report, poverty in 1995 dropped to its lowest level since 1980 both for children and for the overall U.S. population. Because the official definition of poverty does not include non-cash benefit programs or the earned income tax credit, however, this important fact was overlooked when the latest poverty data were released.

The Census data presented in this report also show that the stronger the safety net is, the more people it lifts out of poverty. When the safety net is weakened, it lifts fewer people out of poverty, and poverty rates increase.

The research approach employed here compares poverty rates when the government benefits that families receive and the federal income and payroll taxes they pay are taken into account to poverty rates when these benefits and taxes are ignored. This approach enables us to assess the effect of government programs on poverty.

The analysis also compares poverty rates in different years; for example, it compares poverty rates during the recession years of the early 1980s to poverty rates during the recession of the early 1990s. The analysis examines how differences in the strength of the safety net in these two recession periods contributed to differences in poverty rates during these periods.

These two recession periods provide a particularly useful point of comparison. Reductions were instituted in a number of safety net programs during the early 1980s at the same time the recession took hold. AFDC benefits were ended for substantial numbers of poor working mothers in these years. In addition, food stamp benefits were reduced, a number of individuals were removed from disability rolls, and AFDC benefits eroded sharply in purchasing power.

By contrast, by the time the recession of the early 1990s hit, Congress and the White House had acted on a bipartisan basis to expand the Earned Income Tax Credit, improve food stamp benefits, and extend AFDC to more poor two-parent families. Changes in the Supplemental Security Income program also enanbled more low-income disabled children to receive that program's benefits.

These policy changes affected poverty rates. During each of the two recessions, the number of people who were poor before government benefits are counted increased by about 10 million over the course of the downturn, as the economy deteriorated. But when government benefits are taken into account, the number of people who were poor grew by 11 million during the recession of the early 1980s and by only 5.5 million — just half as much — during the recession of the early 1990s. Poverty grew far less during the more recent recession because the safety net was strengthened, rather than weakened, during the recession itself.

A similar story emerges from a comparison of poverty rates in two years in which the economy was strong \227 1989 and 1995. In 1995, the poverty rate before government benefits are counted was higher than it had been in 1989, the final year before the most recent recession started. But the poverty rate after government benefits are counted was lower in 1995 than in 1989. Expansions in the first half of the 1990s in the Earned Income Tax Credit, the Supplemental Security Income program, and the food stamp program lifted more people out of poverty and drove the poverty rate below its pre-recession level.

Changes in government safety net programs have had a particularly marked impact on poverty among children. Between 1983 and 1995, the number of children removed from poverty by government benefit programs doubled. Not coincidentally, federal expenditures for children increased by 50 percent on a real per capita basis during this period.

The Census data analyzed here also demonstrate that Social Security and other social insurance programs are remarkably successful in shrinking poverty among the elderly. In 1995, government benefit programs reduced the number of poor elderly people by nearly 13 million and reduced the elderly poverty rate from nearly 50 percent before receipt of government benefits to less than 10 percent when government benefits are counted. Stated another way, nearly one of every two elderly people was poor before receit of government benefits in 1995, but fewer than one in 10 remained poor after the benefits are taken into account. Most of this large reduction in poverty among elderly Americans can be attributed to social insurance programs, especially Social Security.

Measuring the extent and depth of poverty as this analysis does can not tell us how people would behave if the safety net programs did not exist. But it shows that a strong correlation exists between policy choices and economic well-being. When the safety net is strengthened, poverty declines. When it is weakened, poverty increases.

These programs have material effects on people's lives. The EITC, to name one program, rewards work by supplementing wages for those who can find only low-paying jobs. Other programs provide temporary help and food assistance for people hurt by cyclical downturns in the economy or provide income support for people rendered unemployable by disabilities.

It is commonplace to say that government failed in the war on poverty. Such a statement is not supported by the data. Poverty does remain high, particularly among minority children, and the welfare system needed reforming. But the fact remains that when Congress and the Administration have worked together on a bipartisan basis to strengthen the safety net, they have eased the depth and severity of poverty and improved economic standards for millions of children and families.

These may be valuable lessons to remember years from now. We are likely to look back at 1995 and 1996 as the time when the safety net for children was strongest. When the poverty data for 1996 are released next year, they are likely to show that further progress in reducing poverty was made this year; economic growth has continued in 1996, and the safety net was strengthened further this year as the EITC expansions enacted in 1993 continued to phase in. is likely the trends found in 1995 will be accentuated. Economic growth has continued in 1996, and the safety net has been strengthened further as the EITC expansions enacted in 1993 have continued to phase in.

In subsequent years, however, the safety net will weaken as the new welfare law takes hold. This law reduces federal safety net expenditures by $54 billion over the next six years and allows states to withdraw an additional $40 billion in state funds from the safety net programs. The data in this report provide powerful evidence that this weakening of the safety net is likely to lead to significant increases in poverty. Such a conclusion also was reached by a widely publicized Urban Institute study issued this summer on the effects of the welfare bill and by Clinton Administration studies of earlier versions of that legislation.

Download the rest of the study. (61Kb)