This year is the 50th anniversary of President Lyndon Johnson’s War on Poverty, and a new paper published by the National Bureau of Economic Research puts us in a far better position to judge how it’s gone. It appears that government programs significantly reduce the number of people living below the poverty line. Yet at the same time, by any global standard, America’s performance in turning economic growth into poverty reduction is incredible—as a negative outlier.
The real value of the U.S official poverty line hasn’t changed much since it was defined in 1963. It’s still meant to be about three times the cost of the “minimum food diet” as defined five decades ago. Broadly, that cost has tracked the consumer price index. What a poverty-line income buys you today, give or take, should look pretty similar to what it could buy you in 1963. That means it doesn’t fully account for the fact that the costs of a lot of vital services, including health care, have grown faster than the general rate of inflation. Nor does it account for changing ideas about what “necessities” are. In 2006 the Pew Research Center reported that 68 percent said a microwave was a necessity and more than half said the same about a computer—two technologies that didn’t exist for consumers in 1963. Thanks not least to falling prices, many poor households now own these devices–81 percent owned a microwave in 2005 and 38 percent a home computer, compared with 88 percent and 68 percent for households overall. But the point remains that a wealthier society thinks more items are “necessary” than a poorer one, and that acts as a force for dissatisfaction and exclusion among those who have a hard time affording them.
That said, the poverty line doesn’t account very well for technological changes or the provision of public services. Take one example: You don’t have to pay for a visit to the library, but today (unlike 1964) the library has more books alongside photocopiers, DVDs, and computers with Internet access. And perhaps most significantly, the official poverty measure doesn’t take account of the impact of government programs like the Supplemental Nutrition Assistance Program, or food stamps, and the earned income tax credit.
A new paper by economist Liana Fox and colleagues calculates historical poverty estimates using the U.S. Census Bureau’s new Supplemental Poverty Measure. That measure accounts for the impact of SNAP, the earned income tax credit, and other government programs on household incomes. It also uses a slightly different definition of the household and measure of the poverty threshold–although one that lines up with the official threshold in 1964 and is only slightly higher than the official line today. Using this supplemental measure, the authors suggest that poverty has declined from 19 percent to 16 percent over the past 50 years. And absent government antipoverty programs, while one-quarter of U.S. households would have been poor in 1967, fully 31 percent would have been poor today.
On the one hand, that’s a real success for the War on Poverty–government programs cut poverty by one-quarter in 1967, they cut it by a half in 2012. On the other hand, these statistics suggest an incredible failure when it comes to America’s ability to convert economic dynamism into significantly improved quality of life and opportunities for the country’s least well-off. The output of the country has climbed from $3.9 trillion to $14.2 trillion. According to the World Bank, allowing for inflation, average gross domestic product per capita in the U.S. has climbed from $17,461 in 1964 to $45,336 in 2012. That’s more than double—an increase of about 159 percent. And yet looking at the numbers on either side of the absolute poverty line, the poorest households in the U.S. appear to have seen regression in pre-program incomes and only a small improvement in post-program outcomes.
From an international perspective, that makes the U.S. an incredible aberration. In 2001, Aart Kraay and David Dollar of the World Bank examined the impact of overall GDP per capita growth on the incomes of the poorest across 90 countries and concluded Growth is Good for the Poor. They suggested that “across regions, time periods, income levels, and growth rates … incomes of the poor rise equi-proportionately with average incomes.” Looking at the income of the bottom fifth of a country across countries and over time, “over 80 percent of the variation in incomes of the poor is due to variation in overall per capita incomes”—leaving just 20 percent to be explained by changes in income distribution. It’s worth noting that a recent paper by Raymundo Campos-Vazquez at El Colegio de Mexico and colleagues suggests Growth is (Really) Good for the (Really) Rich in that the top 1 percent appear to do even better than average from growth across countries. But, with that caveat, Dollar and Kraay’s result have stood the test of time.
And that’s what makes the U.S. look like such an outlier—even more incredible given that it’s a country with a (frayed) safety net, (reasonable) public schools, and (some) infrastructure. Even though American GDP per capita has more than doubled over 50 years, the bottom 15 to 20 percent have seen their income remain pretty much unchanged.
One comparatively benign view of why U.S. economic growth has not reduced pre-program poverty could involve the immigration of poor people. That appears, however, to be only a small part of the story. An analysis of the impact of migration on domestic poverty rates by Steven Raphael and Eugene Smolensky of the University of California at Berkeley suggests that from 1970 to 2005 the change in the population distribution between natives and immigrants added a little more than one percentage point to the official poverty rate. The authors also suggest there’s “little evidence of an effect of immigration on native poverty through immigrant-native labor market competition.” In short, immigrants come to America, earn more here than they did at home, still earn a low enough wage to count as living in poverty, at least in the short term, but have a limited impact on the overall poverty rate in the country.
With the benign interpretation discarded, we’re left with less positive explanations. It appears that as much as the War on Poverty interventions play an increasingly vital role in treating the symptoms of poverty, they don’t adequately address its causes. And these are significant enough to almost completely derail the pattern of economic growth lifting rich and poor alike. Whether the causes are the declining real value of the minimum wage, excessive land use regulation, attacks on unionization, the appalling record of support for job retraining in the face of trade and technology change, health costs, weaknesses in the pension system, or some other set of factors entirely, dealing with America’s abject failure to help poor people help themselves should be an urgent priority for Washington.