Fifty years ago, the democratic socialist activist and writer Michael Harrington published The Other America, which chronicled the state of the American poor at that time. The book shed light on what was at the time an especially marginalized population, easily ignored by middle and upper-class Americans who didn't venture into the urban ghettos and rural communities where most of the poor lived. Most notably, the book, in part due to a review of it by the social critic Dwight MacDonald in The New Yorker, helped convinced the Johnson administration to launch the war on poverty.
The 50th anniversary is prompting a fair amount of useful reflection on poverty reduction efforts since the book's publication, notably at a conference being held at Demos today. Which raises the question: what has happened to poverty since The Other America?
The poverty rate went down, and stayed down
First thing's first: Here's how the poverty rate has changed since 1959, three years before Harrington's book came out:
There are two things to note here. First, there was a huge fall in the poverty rate throughout the 1960s, and in particular after LBJ announced the War on Poverty in 1964 and followed up with Medicaid, Medicare, greater federal housing spending, and other programs to fight that war. In 1964, the poverty rate was 19 percent. Ten years later, it was 11.2 percent, and it has not gone above 15.2 percent any year since then. Contrary to what you may have heard, the best evidence indicates that the War on Poverty made a real and lasting difference.
Second, since the permanent decline achieved during the 1960s, most of the variation in the poverty rate has been cyclical: it goes up in recessions and down during booms. During the early 1980s recession, the rate spiked, only to fall again when the labor market recovered later in the decade. Same thing with the early 1990s recessions and late '90s boom. And the current recession has spurred an increase again. The most recent numbers we have are for 2010, and that year's rate - 15.1 percent - is about as high as it's gotten since the 1960s.
We're doing more to fight it
One of the limitations of the official poverty rate is that it does not count government benefits like food stamps and the Earned Income Tax Credit as income, despite the importance of those programs to the working poor. More recently, the Census Bureau has adopted a metric that takes government programs into account and updates the methodology behind the poverty rate in a number of other ways, such as accounting for regional variations in cost of living and higher medical costs (which can read about in fuller detail here).
That measure is only available for 2009 and 2010, but the National Academy of Sciences proposed a similar alternative measure for which data is available from 1999 onwards, and the Census used a separate alternative metric from 1980 to 2009 as well. Let's compare how the previous Census alternative metric compares to the official one (for the data nerds, I'm using definition 13, as does the University of Michigan's National Poverty Center here):
As you can see, trends in poverty show up the same in both measures, but the alternative measure, which includes Medicaid, the Earned Income Tax Credit, etc., is consistently lower. This tells us that our poverty programs are keeping roughly the same amount of people (1-3 percent of the population) out of poverty each year. In 2009, that number jumped to 2.9 percent. Indeed, Demos estimates that 40 million Americans were kept out of poverty due to food stamps, the Earned Income Tax Credit, Medicaid, and other programs in 2010.
You can also see this if you look at how much the government has spent on anti-poverty programs in the past half century. This chart just includes programs like food stamps, the Earned Income Tax Credit, and Medicaid that solely target low-income people, and even when you adjust for inflation, we're spending far more:
If you include programs like Medicare and Social Security that help many elderly people escape poverty, but also serve middle and upper income people, the rise is even more dramatic:
The bulk of that growth is in health care spending, however, which is growing not because programs are helping more people but because the cost of care is rising, which helps explain why increased spending since the 1970s has not resulted in a sustained decline in poverty.
It's still racially concentrated
Just as in Michael Harrington's day, the poverty rate for black and Hispanic Americans, is much higher than for non-Hispanic whites and Asian-Americans. Unfortunately, the Census didn't distinguish between Hispanics and whites until the early 1970s, and did not start calculating the poverty rate for Asian-Americans until 1987, but even the data we do have are striking:
Both white and black poverty fell dramatically during the 1960s, though the drop in black poverty from 55.1 percent in 1959 to 32.2 percent in 1969 is particularly remarkable. More recently, it appears that white and Asian-American poverty have become decoupled from the state of the economy, whereas black and Hispanic poverty remains heavily cyclical, with a big fall during the 1990s boom and a corresponding rise as the recession hit in 2009 and 2010. Hispanic poverty overtook black poverty from 1994 to 1997, and it looks as though the two are converging again.
UPDATE: A few commenters suggested it might make more sense to look at how government spending has changed in terms of % of GDP, rather than inflation-adjusted dollars, to remove population growth as a factor, as well as moving to a logarithmic scale to highlight the rate of change. Upon reflection, I agree! Here's that chart:
Particularly for low-income programs, the growth in spending during the War on Poverty is quite dramatic, as one would expect. Interestingly, there don't seem to be real cuts during the Reagan years, despite his reputation as a welfare slasher.