Correlation is not causality.
Perry is now associated with AEI - hardly an unbiased outfit. His latest post is an attempt to prove to us, with 2 little graphs, that "government funding increases health care costs" - http://blog.american.com/?p=9051 :-/
The graph you cite doesn't prove anything, I'm sure you know. Employment and unemployment depends on much more than just a bottom-line federal wage rate.
Some say teenage unemployment has risen due to competition from illegal workers - http://www.csmonitor...3-usec.html?s=itm
Some say teenage unemployment has risen due to stay-at-home moms who are looking to supplement their income - http://online.wsj.co...997530518051.html (Note that WSJ article (as opposed to OpEd or Editorial) doesn't say anything about the minimum wage.)
Finally, I think you'll find AngryBear prepares much more persuasive graphs than Perry - http://angrybear.blo...unemployment.html
[...] But if you look at the long term record it is very hard to conclude that the teenage unemployment rate is a function of the minimum wage. Rather, the evidence very strongly implies that the teenage unemployment rate is driven by the business cycle.
[...]
Oh, and take a look at his last graph - http://2.bp.blogspot...h/Clipboard03.jpg
HTH. I think I'm done. :-)
Cheers,
Scott.
(Who thinks that Perry might want to go back to George Mason for some remedial classes...)