In 2013 the U.S. welfare state consisted of three huge programs (Social Security, Medicare, and Medicaid); several other fairly big programs, including Temporary Assistance for Needy Families, food stamps, and the Earned Income Tax Credit; and a number of smaller programs. The Affordable Care Act will eventually become a large program, although not as big as the “big three.” Table 18-3 shows one useful way to categorize the programs existing in 2013, along with the amount spent on each listed program. (The Affordable Care Act was implemented in 2014 so data was not available at time of writing.)

First, the table distinguishes between programs that are means-tested and those that are not. In means-tested programs, benefits are available only to families or individuals whose income and/or wealth falls below some minimum. Basically, means-tested programs are poverty programs designed to help only those with low incomes. By contrast, non-means-tested programs provide their benefits to everyone, although, as we’ll see, they tend in practice to reduce income inequality.

Second, the table distinguishes between programs that provide monetary transfers that beneficiaries can spend as they choose and those that provide in-kind benefits, which are given in the form of goods or services rather than money. As the numbers suggest, in-kind benefits are dominated by Medicare and Medicaid, which pay for health care. We’ll discuss health care in the next section of this chapter. For now, let’s examine the other major programs.

When people use the term welfare, they’re often referring to monetary aid to poor families. The main source of such monetary aid in the United States is Temporary Assistance for Needy Families, or TANF. This program does not aid everyone who is poor; it is available only to poor families with children and only for a limited period of time.

TANF was introduced in the 1990s to replace a highly controversial program known as Aid to Families with Dependent Children, or AFDC. The older program was widely accused of creating perverse incentives for the poor, including encouraging family breakup. Partly as a result of the change in programs, the benefits of modern “welfare” are considerably less generous than those available a generation ago, once the data are adjusted for inflation. Also, TANF contains time limits, so welfare recipients—even single parents—must eventually seek work. As you can see from Table 18-3, TANF is a relatively small part of the modern U.S. welfare state.

Other means-tested programs, though more expensive, are less controversial. The Supplemental Security Income program aids disabled Americans who are unable to work and have no other source of income. The food stamp program or SNAP—officially the Supplemental Nutrition Assistance Program, since it now provides debit cards rather than stamps—helps low-income families and individuals, who can use those debit cards to buy food staples but not other items.

Finally, economists use the term negative income tax for a program that supplements the earnings of low-income working families. The United States has a program known as the Earned Income Tax Credit (EITC), which provides additional income to millions of workers. It has become more generous as traditional welfare has become less generous. Only workers who earn income are eligible for the EITC; over a certain range of incomes, the more a worker earns, the higher the amount of EITC received. That is, the EITC acts as a negative income tax for low-wage workers. In 2013, married couples with two children earning less than $13,430 per year received EITC payments equal to 40% of their earnings. (Payments were slightly lower for single-parent families or workers without children.) The EITC is phased out at higher incomes. As of 2013, the payment ceased at an income of $43,038 for married couples with two children.

Social Security, the largest program in the U.S. welfare state, is a non-means-tested program that guarantees retirement income to qualifying older Americans. It also provides benefits to workers who become disabled and “survivor benefits” to family members of workers who die.

Social Security is supported by a dedicated tax on wages: the Social Security portion of the payroll tax, which was described in Chapter 7, pays for Social Security benefits. The benefits workers receive on retirement depend on their taxable earnings during their working years: the more you earn up to the maximum amount subject to Social Security taxes ($113,700 in 2013), the more you receive in retirement. Benefits are not, however, strictly proportional to earnings. Instead, they’re determined by a formula that gives high earners more than low earners, but with a sliding scale that makes the program relatively more generous for low earners.

Because most seniors don’t receive pensions from their former employers and most don’t own enough assets to provide them with a living, Social Security benefits are an enormously important source of income for them. Fully 60% of Americans 65 and older rely on Social Security for more than half their income, and 20% have no income at all except for Social Security.

Unemployment insurance, although normally a much smaller amount of government transfers than Social Security, is another key social insurance program. It provides workers who lose their jobs with about 35% of their previous salary until they find a new job or until 26 weeks have passed. (Benefits were temporarily extended in response to the severe recession of 2007–2009, with some unemployed workers supported as long as 99 weeks.) Unemployment insurance is financed by a tax on employers; outlays for unemployment insurance were still unusually high in 2013, due to a high national unemployment rate. Like Social Security, unemployment insurance is not means-tested.

Because the people who receive government transfers tend to be different from those who are taxed to pay for those transfers, the U.S. welfare state has the effect of redistributing income from some people to others. Government statisticians have put considerable effort into calculating the effects of this redistribution, which makes a big difference to poverty rates and a somewhat smaller difference to overall inequality. A caveat: such reports calculate only the direct effect of taxes and transfers, without taking into account changes in behavior that the taxes and transfers might cause. For example, they don’t try to estimate how many older Americans who are now retired would still be working if they weren’t receiving Social Security checks. As a result, the estimates are only a partial indicator of the true effects of the welfare state. Nonetheless, the results are striking.

Table 18-4 shows how a number of government programs affected the poverty rate, as measured by the Supplemental Poverty Measure, for the population as a whole and for different age groups in 2012. For each program it shows the amount, in percentage points, by which that group’s poverty rate was reduced by the program. For example, it says that without Social Security, the poverty rate among older Americans would have been almost 40 percentage points higher than it was.

Table 18-5 shows a Congressional Budget Office estimate of the effect of taxes and transfers on the share of aggregate income going to each quintile of the income distribution in 2007 (the latest available date). The effect of government programs was to increase the share of income going to the poorest 80% of the population, especially the share going to the poorest 20%, while reducing the share of income going to the richest 20%.