The latest Newsweek column from Robert J. Samuelson focuses on an issue that deserves a lot more attention than it usually gets: the truth about “poverty.”

As Samuelson reminds us, poverty rates depend entirely on how the people calculating the rates define “poor.” Samuelson also exposes the problems associated with a current Obama administration proposal:

[T]he
administration?s proposal for a ?supplemental poverty measure? in
2011?to complement, not replace, the existing poverty line?… would compound public confusion. It also
raises questions about whether the statistic is tailored to favor a
political agenda.

The ?supplemental measure? ties the poverty
threshold to what the poorest third of Americans spend on food, housing,
clothing, and utilities. The actual threshold?not yet calculated?will
probably be higher than today?s poverty line. Moreover, this definition
has strange consequences. Suppose that all Americans doubled their
income tomorrow, and suppose that their spending on food, clothing,
housing, and utilities also doubled. That would seem to signify less
poverty?but not by the new poverty measure. It wouldn?t decline, because
the poverty threshold would go up as spending went up. Many Americans
would find this weird: people get richer, but ?poverty? stays stuck.

What produces this outcome is a different view of
poverty. The present concept is an absolute one: the poverty threshold
reflects the amount estimated to meet basic needs. By contrast, the new
measure embraces a relative notion of poverty: people are automatically
poor if they?re a given distance from the top, even if their incomes are
increasing.

The new indicator is a ?propaganda device? to
promote income redistribution by showing that poverty is stubborn or
increasing, says the Heritage Foundation?s Robert Rector. He has a
point. The Census Bureau has estimated statistics similar to the
administration?s proposal. In 2008, the traditional poverty rate was
13.2 percent; estimates of the new statistic range up to 17 percent. The
new poverty statistic exceeds the old, and the gap grows larger over
time.