In a recent column to bolster his claim that means-tested programs have only addressed "the symptoms of poverty, not causes", Nick Kristof notes that "the proportion of Americans living beneath the poverty line, 15 percent, is higher than in the late 1960s in the Johnson administration." 

Kristof then goes on to locate the causes of poverty, not in our failure to ensure that wages kept pace with productivity growth over the last several decades, but instead in "a difference in parenting strategies." According to Kristof, the real problem isn't what poverty-expert Sheldon Danziger has described as "the turn to an unequal economy after the 1970s," it's that "working-class families often take a more hands-off attitude to child-raising." 

Conservatives used to think Murphy Brown's childrearing was the problem; Kristof now tells us it's really Roseanne and Dan's parenting that we need to be worried about. 

In a response, CBPP's Bob Greenstein mostly steers clear of parenting practices, and instead takes on Kristof's disparaging of means-tested programs. Greenstein notes that the official poverty measure doesn't count the Earned Income Tax Credit and in-kind nutritional assistance. Although he doesn't say it directly, his implication seems to be that the "real" poverty rate today is much lower than the official Census number of 15 percent. 

It's certainly right to say that millions of children are better off today because of Medicaid, the Earned Income Tax Credit, nutrition assistance, and various other means-tested components of our welfare state. These investments address both the symptoms and the causes of poverty. They also help stabilize the economy during downturns, a function that has broad economic and social benefits. We're a richer nation today as a consequence. 

Still, we have to acknowledge that the poverty rate today isn't lower than it was in the late 1960s. Yes, the official poverty rate (15.1 percent in 2011) doesn't count important benefits, which biases the rate upward. But the official measure also relies on an archaic poverty line that hasn't been updated (except for inflation) for four decades and doesn't take into account the way that broad public consensus about basic needs has evolved since then. This biases the rate downward.  At best, the differences cancel each other out. 

The Census Bureau's Supplemental Poverty Measure includes benefits not counted in the official poverty measure and also updates the poverty line (although too conservatively in my view, especially when it comes to children's basic needs). According to this measure, 16.1 percent of Americans had below-poverty incomes in 2011, a number not that different than the poverty rate in the late 1960s.

Why has income poverty remained high in the United States over the last several decades, despite the expansion of some means-tested benefits? Poverty trends in recent decades have been driven by the same basic economic factors that have battered the broad American middle class. As economist Hillary Hoynes and her colleagues have found:"the lack of improvement in poverty rates … is due to stagnant growth in median wages and increasing inequality."
Another important part of the story is that low- and middle-wage workers are working harder today, but for not much more, than in the late 1970s. As EPI's Larry Mishel finds in a fascinating new paper, the number of hours worked by workers in the bottom fifth of the wage distribution was 22 percent higher in 2007 than in 1979. Middle-wage workers are also working more: 7.5 percent more hours in 2007 than in 1979. At the same time, increases in hourly wages for both groups were miserly: real hourly wages were only 15.8 percent higher over the same period for middle-wage workers, and an especially meagre 7.7 percent higher for bottom-fifth workers. Mishel concludes: "workers have been offering more to the economy and the labor market, and what they have received in return—particularly in the form of real hourly wages—has been very disappointing." 
Along these same lines, I'd add to this that adults living below the poverty line today are much better educated today than in 1979: For example, among middle-aged workers living below the income poverty line, nearly one-third had at least some college or a Bachelor's Degree in 2010, more than twice the share in 1979.
As I've written before, real reductions in poverty will need to tackle inequality and the decline in job quality head on. Maintaining and strengthening existing social insurance programs is an important part of this effort. But just as important is what Yale political scientist Jacob Hacker calls "predistribution": reforming labor and other markets in ways that bring about a fairer distribution of economic power and rewards than we've seen over the last several decades. When it comes to poverty, we especially need to adopt policies that help turn millions of poorly compensated jobs into ones that pay adequately and fairly. 

On this point, Greenstein's response veers a bit off-key when he asserts that "in today's global economy, wages for lower-paying jobs won't likely regain their levels of several decades ago (in inflation-adjusted terms)." Who says? Robert Rubin? Alan Greenspan? Ben Bernanke? Other very serious people who have contributed to the problem through their incredible economic mismanagement and wrong-headed predictions? 

In fact, there's nothing inevitable about wage trends for low- and middle-wage workers, as the responses of Diana Pearce and Don Mathis to the Kristof-Greensteen dialogue make clear. Because of powerful elite interests opposed to a fairer deal for low- and middle-wage workers, the politics aren't easy, just like the politics of the Civil Rights movement or the New Deal weren't easy. But for those of us who care about poverty, there is no alternative: If we don't put policies in place that give low- and middle-income people a much fairer deal in the labor market, poverty will continue to remain high.