You're absolutely right, I've edited my comment slightly. I was making the assumption that total $ robbed was an upper bound on total $ pickpocket, which seems

You're absolutely right, I've edited my comment slightly. I was making the assumption that total $ robbed was an upper bound on total $ pickpocket, which seems reasonable given what I've observed in the United States - but I am having trouble finding reasonable statistics about incidence in the United States.