One of the turning points in the 2012 presidential campaign was Republican nominee Mitt Romney privately saying that 47 percent of the population -- supporters of President Barack Obama, he said -- were people who are "dependent upon government, who believe that they are victims, who believe that government has a responsibility to care for them, who believe that they are entitled to health care, to food, to housing, to you name it."
The 2012 election is long over, but this theme -- sometimes framed as "makers" vs. "takers" -- is an enduring topic in American politics.
Recently, the widely read conservative website TownHall.com posted a column by Terry Jeffrey, editor of the conseravative news service CNSNews.com, in which he compared Americans "on welfare" to those who are "full-time year-round workers." Specifically, Jeffrey wrote, there were more people on welfare than working. He wrote:
"In 2013, according to the Census Bureau, there were 105,862,000 full-time year-round workers in the United States -- including 16,685,000 full-time government workers. These full-time workers were outnumbered by the 109,631,000 whom the Census Bureau says were getting benefits from means-tested federal programs -- e.g. welfare -- as of the fourth quarter of 2012.
"Every American family that pays its own way -- and takes care of its own children whether with one or two incomes -- must subsidize the 109,631,000 on welfare."
A reader asked us to take a closer look at Jeffrey’s column, so we did. We found that the numbers he cited are real, but his descriptions of those numbers, and thus his interpretation, were off base. (Jeffrey did not return an inquiry for this article.)
The number 109,631,000 does, as Jeffrey wrote, refer to the Census Bureau’s figure for the number of Americans receiving benefits from means-tested federal programs. And the Bureau of Labor Statistics’ figure for full-time workers was pretty close to his stated 105,862,000 for 2013, although not all of these workers necessarily worked all year.
But his comparison is a case of apples and oranges. We found three broad concerns about Jeffrey’s characterization, which we’ll take in order.
The term 'welfare' is being used at its most expansive.
The word "welfare" has different meanings for different people. Many think it refers to cash payouts to people who aren’t working; others think it includes anyone who receives government assistance of any type.
Jeffrey has chosen to use the most expansive definition. The number Jeffrey cited includes the "traditional" type of welfare, such as Temporary Assistance to Needy Families, or TANF, but also programs such as Medicaid and food stamps.
In fact, the beneficiaries in the Census Bureau’s accounting who received TANF and other cash-based payments were dwarfed by those who received Medicaid, food stamps and the food program called Women, Infants and Children, or WIC -- the three most widely used categories in the agency’s accounting.
Jeffrey isn’t wrong to frame it that way, but it’s worth noting that the definition he’s used broadens the universe of recipients.
The number includes many children and senior citizens.
Jeffrey doesn’t take into account that the Census Bureau arrived at its figure by counting "anyone residing in a household in which one or more people received benefits from the program." This means the agency would count an entire family of four as receiving means-tested benefits as long as one individual within the family received such benefits.
On its own, this definition is defensible. If one person in a household qualifies for food stamps, then the food purchased with those funds will likely be shared among all family members. However, the problem arises when Jeffrey compares this number to the number of full-time workers.
He’s not comparing the number of households with a means-tested beneficiary to the number of households with a full-time worker. That would have been an apples-to-apples comparison. Rather, he’s comparing the number of households with a means-tested beneficiary to the number of full-time workers.
This matters because it juices the numbers, effectively increasing the number of people on the welfare side of the comparison. Children and senior citizens can be counted as receiving means-tested benefits, but children cannot legally work, and many senior citizens may be retired or physically unable to work due to age.
The difference isn’t trivial. In 2013, more than half of Medicaid recipients -- 51 percent -- were children, and another 5 percent were over 65, according to the Census Bureau. This pattern skews the comparison Jeffrey is trying to make.
The number of 'welfare' recipients actually includes many workers.
Just because you’re receiving means-tested benefits doesn’t mean you’re not working. According to 2012 Census Bureau data, roughly 23 percent of households with at least one working adult received means-tested benefits.
For Medicaid, 28 percent of recipients between the ages of 18 and 64, worked full time, according to the Census Bureau. (Another 15 percent of recipients in that age group worked part time.)
Other means-tested benefits show considerable overlap with the population of working adults. Roughly 60 percent of food stamp recipients who were of working age and weren’t disabled were employed while receiving benefits, according to a Census Bureau sample calculated by the liberal Center for Budget and Policy Priorities.
This matters because working recipients of means-tested benefits would be counted on both sides of the comparison, casting doubt on the notion that there’s a strict divide between people who work and people who are on welfare.
Jeffrey said that in 2013, there were 109,631,000 Americans "on welfare," outnumbering the "105,862,000 full-time year-round workers in the United States."
While the claim is based on real numbers, it’s a fundamentally flawed, apples-and-oranges comparison. The number of "welfare" recipients -- unlike the number of workers -- is enlarged by the inclusion of children and senior citizens. The comparison also ignores that many "welfare" recipients actually work, so trying to separate the two categories creates a false dichotomy. We rate the claim False.