Effective Anti-poverty Programs in the U.S

Effective Anti-poverty Programs in the U.S Hilary Hoynes, University of California, Davis SIEPR Policy Forum on Reducing Global Poverty May 2008 1 Roadmap of talk • Poverty: Definitions • Poverty: Facts • Government policies that reduce poverty (and why) • Government policies that do not reduce poverty (and why) • What else we know about poverty This discussion will emphasize poverty among nonelderly persons in the U.S. with particular attention to families with children. Why? U.S. child poverty rates are high and antipoverty programs for families with children are an area with significant policy attention. 2 (1) Poverty Measurement 3 How poverty is measured in the U.S. • A person is poor if their family income is less than the poverty line (or threshold) • Poverty lines vary by family size and are adjusted for changes in prices each year • Poverty is a family concept—all persons in the same family have the same poverty rate • Income measure is pre-tax family income; only cash government assistance is included in income • This is an absolute measure of poverty (compared to a relative measure of poverty used by many other countries). More on this later. • Definition of poverty has not changed since measure developed in early 1960s 4 Selected Poverty Thresholds, 2007 Household size 1 2 3 4 Poverty Threshold $10,787 $14,291 $16,705 $21,100 Source: U.S. Bureau of the Census For comparison, 2006 median family income was $48,201 for all households and $31,818 among female headed households . 5 (2) Poverty Facts 6 Percent of Persons in Poverty, 1959-2006 25 20 percent in poverty 50,000 40,000 15 30,000 10 20,000 5 In 2006, 34.5 million persons or 12.5% of the population were poor. 10,000 0 0 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 Poverty Rate (all persons) Real GDP per capita 7 GDP per capita (2006 $) Poverty by age group 50 percent of persons in poverty 40 Percent in Poverty, 2006 Children 17.4% 18-64 10.8% Elderly 9.4% 30 20 10 0 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 All Persons 18-64 Children Elderly 8 Poverty by race/ethnicity 50 percent of persons in poverty 40 Percent in Poverty, 2006 White Nonhispanic 8.2% Black 24.3% Hispanic 20.6% 30 20 10 0 1962 1966 1970 1974 1978 1982 1986 1990 1994 1998 2002 2006 All White Nonhispanic Black Hispanic 9 International comparisons of poverty • The U.S. poverty measure is an absolute measure. Most other countries use a relative poverty measure. • [Why? Because in many other developed countries absolute poverty is expected to be very low/zero because of higher spending on redistribution programs.] • Common relative poverty measure: individual is poor if their household income is below 50% of median household income. 10 Relative poverty rate by type of family, 1999-2000 All families 17.0 16.5 12.7 12.4 11.4 8.3 8.0 7.7 7.3 6.5 5.4 10.3 Families with Children 18.8 15.0 15.4 13.2 13.2 7.6 6.0 6.4 9.0 3.8 2.9 10.1 Elderly Families 28.4 48.3 14.4 23.9 6.3 11.2 17.2 17.4 2.0 8.3 10.1 17.0 11 United States Ireland Italy United Kingdom Canada Germany Belguim Austria Netherlands Sweden Finland Overall Average Source: Smeeding (2006), Luxembourg Income Study. Relative poverty rate by type of family, 1999-2000 All families 17.0 16.5 12.7 12.4 11.4 8.3 8.0 7.7 7.3 6.5 5.4 10.3 Families with Children 18.8 15.0 15.4 13.2 13.2 7.6 6.0 6.4 9.0 3.8 2.9 10.1 Elderly Families 28.4 48.3 14.4 23.9 6.3 11.2 17.2 17.4 2.0 8.3 10.1 17.0 12 United States Ireland Italy United Kingdom Canada Germany Belguim Austria Netherlands Sweden Finland Overall Average Source: Smeeding (2006), Luxembourg Income Study. (3) Government policies that reduce poverty (and why) 13 The Earned Income Tax Credit • Refundable tax credit for working, low-income tax filers with children (much smaller credit for childless) • Introduced in 1975; expanded in 1986, 1990, 1993. • Total cost of the EITC is $35 billion compared to $24 billion for TANF and $33 billon for food stamps. • The value of the credit varies with earned income and number of children—with larger credit amounts for families with children (and especially 2+ children). – The average credit in 2006 was $1,806 ($2,693 for taxpayers with two or more children). • Credit available to single parent and two parent families. 14 EITC Benefit Structure, Single mothers in 2007 $5,000 $4,000 Credit Amount (2006$) Flat Region Phase in Region Substantially larger credit for families with 2+ children $3,000 $2,000 Phase out Region $1,000 $0 $0 $5,000 $10,000 $15,000 $20,000 $25,000 $30,000 $35,000 $40,000 Earned Income 2006$ One Child Two or more Children 15 How the EITC reduces poverty 1. Key design feature of EITC (and what distinguishes it from traditional income support programs) is that eligibility requires earned income. – The EITC transfers income to low income families with children WHILE encouraging work. – [At least it encourages entry into the labor market. The fact that the program is eventually ―phased out‖ acts as a new tax for EITC recipients, and in so doing, discourages work. This has not been found to be important in practice.] 2. The generosity of the EITC has increased substantially with tax reforms in 1986, 1990, and 1993. 16 Illustrating how EITC encourages work and reduces poverty $40,000 $30,000 $20,000 Poverty Line, Family of Three $10,000 Earnings $0 $0 $10,000 $20,000 $30,000 $40,000 17 Illustrating how EITC encourages work and reduces poverty $40,000 $30,000 Earnings + EITC $20,000 Poverty Line, Family of Three $10,000 Earnings $0 $0 $10,000 $20,000 $30,000 $40,000 18 EITC payments as percent of cash income by income decile Tabulations of all taxfilers in 2006 10% 9% 8% 7% 6% 5% 4% 3% 2% 1% 0% 1 2 3 4 5 6 7 8 9 10 19 Corresponds to incomes of $5,500 – $17,000 Percent Average EITC Payment, 2006 1 Child $1,715 2+ Children $2,693 Cash Income Decile (among all taxpayers) Real EITC Benefits Increasing over Time (2+ children) $5,000 2006 [After OBRA93] 1993 [After OBRA90] $4,000 1988 [After TRA84] 1984 [Before TRA86] $3,000 Credit Amount (2006$) $2,000 $1,000 $0 $0 $5,000 $10,000 $15,000 $20,000 $25,000 $30,000 $35,000 $40,000 Earnings 2006$ 20 • So, the EITC has the potential to increase incomes and reduce poverty through two channels: – The EITC represents an increase in income for the family – The EITC provides incentives to enter work, and thus increase earnings which increases family income. • Many studies have documented that indeed the EITC is a key factor in contributing to the observed increases in employment rates of single mothers with children. • As an illustration … 21 Percent of Women Working (by Marital Status and Children) 100% 95% Percent employed at all last year 90% 85% 80% 75% 70% 65% 60% Single, No Children Married, No Children Single, Children Married, Children 1983 1986 1989 1992 1995 1998 2001 2004 On the eve of these policy changes, much attention is given to low employment rates of single mothers. 22 Percent of Women Working (by Marital Status and Children) 100% 95% Percent employed at all last year 90% 85% 80% 75% 70% 65% 60% Single, No Children Married, No Children Single, Children Married, Children 1983 1986 1989 1992 1995 1998 2001 2004 Beginning in 1992—dramatic increases in employment for single mothers, with little change for other women 23 Census estimates of poverty reduction due to EITC, 2003 2 Not including EITC 35.9 EITC lifts 4.3 million persons out of poverty 1 Including EITC 31.6 0 10 20 30 40 50 24 Number of persons in poverty (millions) (4) Government policies that do not reduce poverty (and why) 25 Cash Welfare Programs (TANF) Income support (welfare) programs are unlikely to affect poverty rates for two reasons: 1. Benefit levels are very low, unlikely to increase a household’s income over the poverty line. 2. Benefits are targeted on those out of work; thereby discouraging work rather than encouraging it. [This does not mean the program is not important or useful. Rather that it simply should not affect poverty rates.] 26 Illustrating why TANF will not reduce poverty $40,000 Earnings $30,000 $20,000 Poverty Line, Family of Three $10,000 $0 $0 $10,000 $20,000 $30,000 $40,000 27 Illustrating why TANF will not reduce poverty $40,000 Earnings $30,000 $20,000 Poverty Line, Family of Three $10,000 Earnings + TANF $0 $0 $10,000 $20,000 $30,000 $40,000 28 One could use TANF to reduce poverty, but funding would have to be about four times the current funding. $40,000 Earnings $30,000 $20,000 Poverty Line, Family of Three $10,000 Earnings + TANF $0 $0 $10,000 $20,000 $30,000 $40,000 29 Expansion of EITC and Contraction of Welfare has Radically Changed the Landscape for Low Income Families with Children $50 $40 $2006 Billions $30 $20 $10 AFDC/TANF Expenditure EITC Expenditure $0 1975 1980 1985 1990 1995 2000 2005 30 Other things you should know about poverty • U.S. official poverty statistics are defined using pre-tax cash income. So the EITC does not affect official poverty. Neither do the many non-cash assistance programs (food stamps, housing subsidies) • Minimum wages are not very important. • The strength of the low skill labor market more generally matters a lot. The late 1990s expansion raised earnings at the bottom of the skill distribution which led to decreases in poverty.5 • Increases in Social Security benefits are the main factor explaining the historical reduction in poverty rates among the elderly. • Changes in immigration are not a major contributor. 31