The Census’ Supplemental Poverty Measure paints a different picture of the poor and the social safety net.

In 1987, President Ronald Reagan announced defeat at the hands of the “war on poverty“—a war his predecessor Lyndon Johnson had waged since 1967. InThe Atlantic the year after, Nicholas Lemann explained why all anti-poverty programs need not be as “ill-fated” as the ones spearheaded over those decades.

But were they really as ill-fated as he observed? What about the government assistance programs that followed—how effective were they? The answer to those question lies in a new(ish) measure the Census Bureau uses to measure poverty.

Last week, the bureau released the poverty rate based on the Supplemental Poverty Measure. This picture of the poor in America is slightly different from the one released last month, based on the standard official measure.

The 2013 poverty rate based on the supplemental measure is 15.5 percent—higher than the official 14.5 percent rate. In real numbers, that translates to 3.4 million Americans living below the poverty line in one universe (measured using the SPM), while living above it in the other (measured using the OPM).

The supplemental measure uses more updated formulas associated with consumption patterns and family structures to calculate the rate. It also subtracts essential costs—food, clothing, housing, and utilities—and adds the non-cash benefits provided by the government. Arguably, it’s a more comprehensive measure.

“[The SPM] tells us a lot more about how poor people actually are,” says Isabel Sawhill, a senior fellow at Brookings. “It includes resources they have access to and costs that they incur.”

The measure was born in 2009, following from the 1995 recommendations by the National Academy of Sciences. The Census Bureau released the first set of estimates in 2011. Because the data go back only to 2010, the measure hadn’t been stretched back to examine historical poverty trends.

That was until earlier this year, when researchers at Columbia University decided to take a crack at it. Using the Consumer Expenditure Survey and theCurrent Population Survey, they generated the supplemental measure-based poverty rate for the past 50 years. The analysis puts the efficacy of anti-poverty and government assistance programs in combating poverty in a new light.

“Did [the programs] accomplish nothing or did they accomplish a lot?” Sawhill asks. “Well, the answer depends on which measure you look at.”

If that measure is the official one, the overall poverty rate over these decades hovers around 15 percent—hardly changing. If we look at the supplemental rate, it takes a dive over those decades (19 to 16 percent).