Arnold Kling points out a very interesting CBO study released last week on the progress of low-income families with children. I am in particular fascinated by the results presented from the 2001 Survey of Income and Program Participation (SIPP) which allowed tracking the same families that were low-income in 2001 through the following three years.

The fact that average income was increasing between 2001 and 2003 for those households with children that were in the low-income category in 2001 does not mean that income was increasing for every household. Although about 60 percent of female-headed and other low-income households with children in 2001 experienced real increases in income of more than $1,500 over the next two years, about 25 percent of each group saw their income decline by more than $1,500 over the same period (see Figure 9).18 For households in which income increased, it did so substantially, nearly doubling between 2001 and 2003.

A $1500 increase is an increase of about 10%. The changes were not due to changes in marital conditions of these families. Also take a look at Box 1 of the report (p. 9) in which they show that only half of the families making the increases had bounced back from a bout of unemployment.

Kling notes that while we perceive no change in poverty rates, the income of the lowest quintile of families with children increased 35% between 1991 and 2005. Only for the top group did it grow faster. That does seem counterintuitive to me, as Kling points out, the effect of immigration on the poverty level should be to hold its income growth down. Or maybe that story just isn't right.

Labels: economics