Half empty or half full?
You decide.
This is from the Pew Economic Mobility Project. Look at it carefully. I'll try to say more about it later and to bring some additional data.
It is from the PSID. It follows the same people. It is corrected imperfectly for inflation. But it's the best you can do.
Podcast RSS Feed
Full EconTalk Text
{ 7 comments }
Speaking of following people, here's a chart showing how bachelor degree earners can expect their incomes to change, as a percentage of their "starting" income, over the first 25 years after they leave school.
The anomaly, 2001, was a year in which the average starting income for new college graduates was unusually high – this is an artifact of the recession of that year (basically, the graduates in certain highly demanded fields skewed the statistics as fewer of those graduating in lesser demanded fields found work.)
The percentages alone are not enough to blow me away. Yes, it is impressive that the majority of children that were born to mid to lower quintile parents did better than their parents. What would be helpful is to find out how much better they did and what percentage ended up in what quintile. I am really surprised that the percentage of children from the top quintile is as high as it is; I had assumed that even given what a troll might call "the bettre environment to be born into", that more chilren in those situations wouldn't meet the income-earning potential of their very suceessful parents. I have this image of many children of successful families letting their parents down, adopting destructive habits, becoming slackers, or, best case scenario, not have the same drive or talents.
Russ, I think your link might be broken.
There is the expected regression to the mean, so I'm not really sure how to take these results. Wouldn't it be more informative to follow individuals' incomes over time rather than comparing parents to offspring?
The link is broken, and it is very long.
Here's a short version:
http://alturl.com/4jfm
The family composition chart on page 3 seems very relevant.
How come the overall numbers are very similar between blacks and whites, but when you break it down by quintiles the disparity is much wider? Is there some stat magic here?