Here's a very recent study on mortgage default rates, and the differing modes of default, between recourse states and non-recourse states.
It's a .pdf file, and I'm not smart enough to know how to cut/paste from it. Major conclusions:
1) (No surprise): Default is less likely in recourse states. "At the mean value of the default option at the time of default, the probability of default is 20% higher in states with no recourse as compared to states that allow recourse." Also, the number of deficiency judgments probably understates the importance of recourse laws because the threat of a deficiency judgment probably deters many people from skipping out when they are underwater. (the term "strategic default" is used in the paper).
2) In recourse states, when defaults do occur they happen in ways that result in reduced losses to the lenders (again, no surprise).
3) In recourse states, wealthier people are less likely to strategically default than those with less wealth.
Of interest: "The no-recourse moertgage is virtually unique to the United States. That's why falling house prices in Europe do not trigger defaults"
I was hoping to find a study comparing loan costs in recourse vs. non-recourse states. If this info is in the study linked above, I didn't see it. Whatever the difference in rates/terms were before the "bubble", I'll bet going forward there is more of a difference.
IMO, both options (recourse loans and non-recourse loans) should be available on the market to borrowers, if sellers want to offer them. Then the actual cost of the "jingle-mail" option will be explicit to everyone. And, there'll be less stigma from walking away from a mortgage. Some will see that as a bad thing, but I'd rather that the entire bargain be explicit so that people with an (overly?) developed conscience aren't paying a bill that they don't need to, and so those with less scruples aren't getting an advantage.