As I said, if you start out with a flawed list, you're going to end up with a flawed list. If you draw jury lists from municipal property assessment lists, you start out with a list that represents only a fraction of the population. There's very good evidence.
The Toscano Roccamo report, which Mr. Friedman referred to, provides empirical evidence that if you start with a property assessment list to draw your jury lists, you will end up with jury panels composed predominantly of middle-class, white homeowners. There is data on that. My point here is that if you start with a flawed list, you'll end up with a flawed jury panel, and you cannot resuscitate that flawed list by exercising peremptory challenges.
In any event, even if there is a marginal benefit from the standpoint of equality or diversity, you have to measure that against the downside of having these challenges and the potential for their misuse. I think there is lots of scholarship to suggest that the potential for misuse is real.