At 6 feet 3 inches and 250 pounds, President Lyndon Johnson was an oversized man famous for getting his oversized nose right up in another man’s face and badgering him with data, stories and sheer energy until he gave in.
This persuasive assault, known as “the treatment,” was usually a one-on-one affair. But when Johnson stood before Congress on Jan. 8, 1964, to declare a “war on poverty” the whole country was set back on its heels.
“This administration today, here and now, declares unconditional war on poverty in America,” Johnson said. “It will not be a short or easy struggle, and no single weapon or strategy will suffice, but we shall not rest until the war is won.”
“Our aim,” he added, “is not only to relieve the symptoms of poverty, but to cure it, and above all to prevent it.”
Republicans were flummoxed. “It’s the Sermon on the Mount,” one Democrat told the Washington Post. “How can anyone attack it?”
A flurry of legislation followed. Some programs, like Medicare and Medicaid, had lasting impact. But as the Vietnam War consumed the Johnson Presidency, federal focus and resources were diverted from the War on Poverty.
More to the point, it soon became clear that both wars lacked strategic vision. Declaring war was one thing. Knowing how to win it was another.
“My friends,” said President Ronald Reagan twenty-four years later in another State of the Union address, “some years ago, the federal government declared war on poverty, and poverty won.”
Using the official poverty rate, Reagan had a point. And it still holds.
Poverty in 2010 was 2 points higher than in 1970, Bruce Meyer and James Sullivan noted in a recent Brookings paper, despite years of economic growth and trillions of anti-poverty spending. And by 2011, 15 percent of Americans stood below the line, up sharply from 12 percent a decade earlier.
And yet, the numbers are murkier than they sound.
For nearly 20 years, experts have debated what it means to be poor. Finally, in the fall of 2012, the Census Bureau for the first time released its long-awaited Supplemental Poverty Measure, which some think more accurately measures true poverty.
How we measure poverty has weighty policy implications. For starters, the poverty line shapes how policy makers and the public think about various disadvantaged groups, and how we measure policy success.
The new supplementary measure, for example, counts more elderly as poor by placing more emphasis on medical costs. A third alternative, proposed by Meyer and Sullivan, finds fewer poor overall, but more of them are poorly educated, Hispanic and married couples with children.
Policies grow directly out of these assumptions. The new health care law, for example, guarantees free care to those near the poverty line, currently $23,050 annually for a family of four. Scaled subsidies are then given up to 400 percent above that line.
The new measure now appears alongside the official poverty line, but does not replace it. Not yet, anyway. In time, the two competing standards may flip places. Or both may be supplanted by a third. One thing is clear: for the first time in fifty years, what it means to be poor in America is up for grabs.
A simple yardstick
The official poverty measure is the handiwork of Molly Orshanksy, the economist who developed it for the Social Security Administration in 1963. Except for cost-of-living hikes, her poverty standard has remained unchanged for the past fifty years.
Noting that typical Americans at the time paid roughly 1/3 of their income on food, Orshansky simply took the cost of a nutritious diet and multiplied it by three to account for other expenses.
“It was quite an ingenious concoction considering how few ingredients she had in the statistical kitchen,” said Nicholas Eberstadt, an economist at the American Enterprise Institute. “It’s kind of like a glorified all-nighter in college.”
Most agree this approach was reasonable at the time. “But nowadays Americans are spending scarcely 10 percent of their income on food,” said Eberstadt. By failing to capture the rise of payroll taxes, transportation, childcare and out-of-pocket health care, critics argue, the official poverty line understates burdens on poor households.
On the other hand, the old standard also ignores help offered by government anti-poverty programs. Food stamps, earned income tax credits, childcare credits and housing subsidies do not count in measuring poverty.
“So we are not really looking how successful the War on Poverty has been,” Eberstadt said.
In short, the 50-year-old poverty measure understates costs but also understates government assistance. The two balance out in a rough way.
Bruce Meyer argues that the rough balance still grossly understates the progress that has been made. “What the official measure says,” Meyer said, “is that poverty is worse today than it was in 1981. And that's nonsense.”
In a curious political alignment, both conservatives opposed to anti-poverty programs and liberals who want to bolster them try to downplay success in fighting poverty, according to Meyer. “The majority view among poverty researcher colleagues seems to be if we don't say poverty is worse today than it was 30 years ago people will not be interested in supporting antipoverty efforts."
“If you believe those numbers,” echoed Nicholas Eberstadt, “that suggests that over 50 years of pretty steady economic growth we have made no progress for the poorest strata of society.” Eberstadt, like Meyer, calls this “nonsense.”
Meyer notes that in 1981, just 15 percent of Americans had central air and only 41 percent had either window units or central air. By 2009, with the official poverty rate actually higher than in 1981, 55 percent had central air and 83 percent had some form of air conditioning. “To be poor today in the U.S. is nothing like what it was to be poor 40 years ago,” Meyer said.
One key question is whether poverty should be measured as an absolute standard of need, or whether it should be calculated relative to what others enjoy. David Betson argues the latter.
“We wanted to get away from what experts believe people should be consuming and look more at what American people actually are doing,” said Betson, a Notre Dame public policy professor who helped develop the new supplementary poverty measure.
The National Academy of Sciences panel that framed the new measure decided that the “needs of a person in a population are relative to what the consumption of that population is," Betson said. "So as people spend more, to participate in that society, they will actually need more. As the society gets wealthier, they need more income.”
Thus, the new poverty standard uses a complex formula that ties the needs of the poor to income levels of the general population. If Americans get richer, so do the poor.
“It depends on how you view what poverty means,” Betson said, admitting that this formula is a “debatable point” and a “bone of contention.”
Eberstadt would like to pick that bone, arguing that under Betson's method “poverty rates would fall if the country as a whole became less wealthy.”
Betson conceded the point but would not apologize. “If we as a nation over the long run became less wealthy, then the answer would be yes, needs would start to fall — in real terms," Betson said.
Simply put, in Betson's vision, the poor in a richer nation are richer than the poor in a poorer nation.
What to measure
Setting aside what people actually need, measuring their resources is also a contentious matter. While both the official and supplementary measures focus on income, for example, some argue that it makes more sense to look at what people actually consume.
Someone who has a bad year in sales or who goes back to school for a year or two may survive a lean year by consuming savings or by using credit. For them, Meyer said, “one year’s income is a noisy measure of their life situation.”
To reduce this noise, Meyer and Sullivan use government surveys to track household spending. They argue their model is better at catching the real poor and filtering out those who are not.
Changing the formula changes the outcomes. Consumption-based poor had much smaller homes than either of the two income-based poor. They were also much less likely to have finished high school and were also much more likely to be Hispanic.
On the other hand, the new supplementary measure finds more elderly poor because it factors in medical costs, which grow with age, and because it emphasizes current income over consumption of existing assets.
Betson thinks the leap in elderly poor in the SPM simply reflects reality. “How can you have a measure that doesn’t account for medical care?” he asked. “If you are ignoring medical care in your poverty measure, the elderly are going to look pretty good. You are ignoring the largest component of their needs.”
On the other side is Meyer, who argues that the elderly have more resources than income measures capture.
“With the elderly, if you just focus on their income you will miss that many of the elderly have substantial assets that they can draw down,” Meyer said, noting that more than 80 percent of those 65 or over own their own home. They may not own those homes outright, he acknowledges, “but if you bought a house 20 years ago, the mortgage is probably much lower than rent equivalent.”
Married with children
All three models are fairly close on the poverty rates among single-parent households, but they differ sharply on married couples with children.
Thirty-eight percent of poor households in the consumption model are married with children, compared to just 25 percent in the new supplementary measure (SPM). The official Orshansky model splits the difference at 32 percent.
Meyer is not quite sure how to account for this gap. One possibility is that anti-poverty programs systematically favor single parents over married families. Another might be that married families are more likely to resist assistance and muddle through on their own. Whatever the explanation, Meyer said, “these numbers do suggest we should pay attention more for how our programs affect married couples.”9 comments on this story
This is one place where, as Betson argues, differences between competing models may actually shed light on issues that would have remained hidden had any one measurement held sway.
In a response to Meyer and Sullivan, Betson and coauthor Constance Citro argued that we may be able to “make use of the strengths of each type of measure to enrich policy analysis and public understanding of the extent and condition of the poor in the United States.”
In any case, it seems clear that a half-century of unchallenged dominance enjoyed by the old poverty formula is rapidly ending.