The title of this post should have been the title of my post yesterday. As one of the commenters, Zach Pruckowski, pointed out, the new poverty measure that the U.S. Census Bureau uses does include food stamps. My bad.

But here’s what else the U.S. Census Bureau’s new Supplemental Poverty Measure, introduced last month, does: it changes the way poverty is measured from an absolute measure to a relative measure. Here’s the Census Bureau’s explanation of the “weakness” in the old measure:

The current poverty thresholds do not adjust for rising levels and standards of living that have occurred since 1965. The official thresholds were approximately equal to half of median income in 1963-64. By 1992, one half median income had increased to more than 120 percent of the official threshold. [italics in original]

What’s the Census Bureau’s fix? Make poverty a moving target. If standards of living improve, as most of us optimistic economists expect, then define as poor those people who are below some percentile. Here’s the Census Bureau’s specific statement:

Adjustments to thresholds should be made over time to reflect real change in expenditures on this basic bundle of goods at the 33rd percentile of the expenditure distribution.

So now throw in now just poor people but also low-income people, with low-income also defined relative to others, and you’re just about at the median, as commenter Chris T writes. What the news story I reported on yesterday essentially reported is that half of American households are below the median. Which we already knew.

Oh, and my apologies to Sheldon Danziger. I, not he, am the one who was unaware of the new measure.

HT to Tim Worstall.