Despite a year in which the US economy added jobs, the percentage of Americans living in poverty grew from 12.5 to 12.7 percent last year - the fourth straight year it's risen.
That increase, reported in the much-anticipated annual Census Bureau study Tuesday, surprised many analysts who had expected the number to drop along with unemployment.
Political pundits on both sides of the aisle rushed to put their stamp on the news, with Democrats blaming the trend on failed economic policies and Republicans pointing out that some regions and groups improved.
While the overall poverty numbers went up, for instance, the Midwest was the only region that experienced an increase. Rates for child poverty and the uninsured were unchanged - after experiencing a rise the year before - and most measures of income gap showed no change.
While the means of calculating the statistics have drawn criticism from both the right and the left, many experts see the annual figures as a useful yardstick by which to measure progress over time. And for some, the lack of long-term improvement is particularly troubling.
"There is still a generation of no progress against poverty," says Sheldon Danziger, codirector of the National Poverty Center at the University of Michigan. "Somehow, we have to confront the fact that ... a rising economy no longer lifts all boats."
Others, however, were optimistic.
"America looks like a giant jobs machine still," says Douglas Besharov, director of the American Enterprise Institute's Social and Individual Responsibility Project. "Sure, it'd be nice if we got out of recessions faster, but this is a very firm base from which to build."
Mr. Besharov and other conservatives point to the fact that over the last several decades, progress has been made on quality-of-life factors such as housing quality and life expectancy of the poor. And they emphasize that this most recent rise in poverty seems limited to the Midwest.
"It's not uncommon for poverty to go up three or four years after a recession is over," says Kirk Johnson, a senior policy analyst at the Heritage Foundation. "And this slight increase in poverty is being driven by what is going on in America's heartland, rather than anything systemic throughout the country."
The numbers show a plateau for many of the indicators the report measures. While the number of those in poverty increased by 1.1 million from 2003 to 2004, to a total of 37 million, Asians were the only group whose poverty rate declined, while non-Hispanic whites were the only group to show an increase.
Some analysts say the numbers are particularly troubling not so much for the change they show, but the lack of it - the persistance of relatively high poverty rates over time. The real-life picture for those under the poverty level - or even many above it - can be grim, they say.
"A lot of us are very concerned that it seems increasingly difficult for poor children to achieve the kind of mobility that generations prior to ours had," says Margy Waller, a Brookings Institution scholar who served in the Clinton administration.
She and others say that poverty could be reduced if the government were to create and commit to more programs that focus on the working poor. In 2003, for example, despite a spike in child poverty, the percentage of children without health insurance remained steady from the year before. Many cite efforts to expand Medicaid enrollment and the State Children's Health Insurance Program as significant factors.
The US has made fewer strides in reducing poverty, critics say, than other industrialized nations. England, for example, has been cited for successfully reducing child poverty. "[Prime Minister] Tony Blair did in the UK in 1999 what President [Lyndon] Johnson did here in 1964, that is to say, 'We are going to make fighting poverty a priority,'" says Dr. Danziger.
David Brady, a sociology professor at Duke University, says that part of the problem is that the poverty level itself is far too low - that many above the threshold still do not have the means to make ends meet. He says no president has had the incentive to acknowledge that the level is too low, and the American public has not demanded accountability. "We don't care enough about the poor," says .
The data comes amid growing criticism that the statistics - or the means of determining them - are faulty. While the world has changed, say critics on both the right and the left, the formula for determining poverty in that world has not. Some claim the official poverty level underestimates the number of Americans who are poor, while others contend the opposite.
Critics have assailed the annual figures because they are not adjusted for geographical standards of living, nor do they account for the rising significance of housing, health, and childcare costs. Developed in the 1960s, the poverty level is determined by the amount required to feed a family. But while food has become a less important expenditure for today's families, costs for things like childcare have gone up.
"A poor family then looks very different than it does today," says Dr. Brady. Conservatives, meanwhile, point out that the rate might be inflated because it doesn't include non-cash benefits like food stamps or housing assistance.
Among the most contentious elements of the report is whether it accurately reflects the number of uninsured in the US. Many say the Census Bureau undercounts those participating in Medicaid, whose enrollment has expanded significantly under President Bush. According to two independent studies from The Urban Institute and the Actuarial Research Corporation, the Census has overestimated the number of Americans without insurance in a full year by anywhere from 4 to 9 million people.
"It's like using a broken yardstick," says Mr. Besharov. "But if you use the same one year after year, it tells you something."