The point is that since those incarcerated previously made up a significant fraction of the bottom end of income distribution, removing them from the pool (i.e. throwing them in the slammer) has the effect of making p.c. income increase. What I'm suggesting as a possibility (guidance on this point by "the pros" greatly appreciated) is that those towards the top and at the middle are doing about the same and those at the bottom are simply treated as if they don't exist for the purpose of statistics. Not my idea of a big improvement, but the sort of "improvement" which Clinton routinely trumpets, hence the remark below.
> >In general, when it comes to economic statistics, you've got to
> >continually check your wallet and count your change when your dealing with
> >these guys, it seems to me. (The NAFTA fight demonstrated this in spades.)
>
> You can't trust anything coming out of the White House, but the
> Census Bureau is about as reliable as they get.
>
Right. It's the interpretation of the numbers-how they're spun-that counts.
Btw, on the subject of the quality of Goodman's questions, my understanding is that she received a call from the White House that Clinton would appear on the air in five minutes, breaking into a latin music show. Not a whole lot of time to prepare. I don't imagine any one of us would have done a whole lot better.
Judging from Clinton's response, she didn't do all that badly.
John