Understanding Poverty: What the Census Doesn't Tell Us

It’s time we change how we think about poverty. The newly released Census report on poverty received a lot of attention from the chattering class. But was it really deserved? 

There are many ways in which the rate understates poverty. The poverty line, individuals making $11,484 a year, has been used since 1964. A CBS report explores its inadequacy: 

This amount was originally set as the earnings line below which a family of three or more would have to spend more than a third of its income on food. Notably, the report does not consider other major expenses, such as housing, transportation, medical care and child care, in gauging the nation's poverty rate.

That’s a big blind spot. 

In his article for the Prospect’s Poverty issue, Mark Levinson provided a thorough critique of the way that the Census measures poverty. 

The poverty line was roughly equal to 50 percent of median income when it was set in the 1960s. It has now declined to about 36 percent. So today’s official poor are further behind the average income—in a word, poorer—than their counterparts were 50 years ago. If we measured the number of Americans who are below 50 percent of the median income, 22.6 percent (69.1 million) are poor—nearly half again as high as the official poverty rate of 15.1 percent. 

Levinson goes on to attack the Census’ other measurement, which will be released in November, that accounts for programs like food stamps and unemployment insurance, called the Supplemental Poverty Measure (SMP). SMP may incorporate government programs, but it abides by the same misleading poverty baseline as the traditional Census reading. It too is outdated.

In addition, while we can ascertain trends from the Census numbers, it doesn’t account for transfer credits. The headline number -- 15 percent of Americans living in poverty -- is based solely on income. It doesn’t measure the impact of government programs designed to actually reduce poverty, like food stamps, unemployment insurance or the Earned Income Tax Credit.

 

The Brookings Institution released a report yesterday that incorporates the effects of government programs, measuring consumption rather than income. Their conclusion? 

Changes in tax policy explain a substantial part of the decline in income poverty particularly for families with children. Other than social security, cash and noncash government transfer programs have only a small impact on changes in poverty

It's interesting that Social Security had by far the biggest effect of any social program, far greater than TANF, the most prominent anti-poverty program. Further, it's striking that tax policy, particularly the Earned Income Tax Credit, has had a greater impact than any government transfer program, including Social Security.

This exposes a tension at the heart of American governance: the preference for complicated tax credits rather than direct spending programs. When policies like food stamps, unemployment insurance, and TANF are underfunded, they just don't provide much relief to the poor. Whereas huge programs like the Earned Income Tax Credit do, because they're classified as tax breaks rather than spending.

Whether they understate or overstate the extent of poverty it’s clear the headline Census numbers don’t tell us much. Further, our programs ostensibly combatting poverty do much less than most Americans think. Derek Thompson at the Atlantic interprets the Brookings study as evidence that we're winning the War on Poverty. I would argue that Brookings shows we're fighting the wrong battles in the war on poverty. Clear-sighted, well-funded programs can have a big effect. Unfortunately, our effective programs are underfunded or too constrained by the federal government's narrow definition of who is poor.

It’s no coincidence that the solutions are inadequate if the measurements are unclear. 

Not only do we need to redouble our commitment to fighting hardship, we badly need to better understand it. Our peers confront poverty with more accurate assessments, more strident solutions, and, as a result, more equality. It's time we do the same.

Comments