Speaking of the crime issue, I should say that while I think the United States has gotten unduly complacent about crime since the big "crime drop" of the 1990s. There's no reason we shouldn't be able to get the murder rate back down to 1950s levels if we try, and I think there's pretty good reason to believe that sustained further reductions in crime levels can help create a situation where our ludicrously high prison population can start coming down. What's more, I think we actually learned a lot during the 1980s and 1990s about effective crime control strategies. McMegan waxes vague:
On the other hand, "culture matters" doesn't get you very far as a poverty eradication program; no matter how much money you give welfare mothers, they'll still be on welfare. And "they're poor" has proven to offer little in the way of crime-reduction strategies; we've been much more successful with things like more police on the beat.
And, indeed, there is good evidence that more police officers helps reduce crime. What's more, though I really, really, really don't think Rudy Giuliani should be elected president, the Compstat system he and William Bratton implemented appears to be a helpful way of deploying the cops you have more effectively. And, indeed, Bratton is currently following up his successful runs in Boston and NYC with successes in Los Angeles and all this presumably has something to do with his methods (i.e., things that can be copied) rather than with his Batman-style crime fighting abilities.
What's more, as I somewhat reluctantly concede in this article, mass incarceration, though cruel and expensive, is somewhat effective as a crime-control strategy. Meanwhile, another big part of the 1990s story is that while welfare handouts may not reduce crime (this is a bit unclear, since welfare goes overwhelmingly to women and crime is committed overwhelmingly by men -- the gender gap also being a good reason to doubt that objective material need is the key variable here) wages for unskilled workers are an important determinant of crime rates in a way that's pretty intuitive -- if there's more money to be made working, you're ceteris paribus more likely to spend time working, and when you're behind the desk at CVS you're by definition not robbing anyone.
Relatedly, while I don't really believe that improved educational attainment is the key factor in curbing inequality, better performance at the bottom end -- fewer high-school drop-outs in particular -- could do a lot to reduce crime. Similarly, high-intensity early childhood interventions seem to work.
Last, while we spend a fortune on locking people up in prison, we also have all these parolees who are being supervised in a very bare-bones and largely useless manner. Something more expensive than traditional parole -- something like "Coerced Abstinence" -- would be way cheaper than full-time incarceration but way more effective than what we're doing now.
Photo by Flickr user Hisgett used under a Creative Commons license