Dylan Mathews on the Census Bureau's "supplementary poverty measure," which is intended to overcome some of the shortcomings of the traditional measure of poverty:
The new poverty measure is out, and it’s grim, by Dylan Matthews: ...In recent years the Census Bureau has begun developing a “supplemental poverty measure”... Today, it released the supplemental figure for 2011. Overall, it’s higher than the official measure, at 16.1 percent, but for some groups, such as children under 18 and blacks, it’s actually lower. By contrast, it’s much higher for the elderly (15.1 percent in the supplemental measure, 8.7 percent in the official one)...
Perhaps the most interesting part of the report is the Census’ measurement of how much various government programs and categories of expenses reduce or increase the supplemental poverty rate, which unlike the official rate, the supplemental measure takes into account. Medical expenses are the main expense contributor to poverty, followed by expenses related to work (such as transportation, supplies, etc.), while Social Security is far and away the most important program for reducing poverty, followed by tax credits like the Earned Income Tax Credit (EITC) or the child tax credit (CTC):
...the Social Security number is especially notable given how much higher the supplemental measure is than the official one for the elderly. It suggests that even with that substantial safety net, the poverty problem among the elderly is much bigger than we thought.