I've read about a couple of different explanations for changes in crime rates. I think freakanomics has discussed greater prevalence of abortions[1], and one thing I've heard in academic circles is that crime reporting has gotten better. You could actually make predictions with this article's question. There may have been areas where lead was used in petrol for longer than others, one could compare across regions and see if the time at which petrol was changed also adheres to drop in crime to see if there is still a correlation.
http://freakonomics.com/2005/05/15/abortion-and-crime-who-sh...