I guess, I'd probably reference you, Joanne, back to Page 15 in our discussion this morning. What that really shows you is -- just articulated again is what the delinquent risk that is currently in some form of foreclosure, the levels that we have that reserved against as a percentage of our overall risk in-force. In the quarter, you continue to have some uptick in foreclosure development and in a couple of the sand states. But overall, at the same time, it's just our view that these are going to be harder to cure. You're going to have less loss mitigation opportunities in those compared to where we were early on in the cycle. And if you think back on it, as we went into 2009 and we saw an acceleration in our loss mitigation benefits that was coming from both rescission activity that was really geared towards a lot of the problem loans and the problem loan types that were originated in those sand states, coupled with the growth in loan modification benefit that we really saw across the country, and in the sand states as well. We reflected that in our reserve calculations. And what we've experienced, really, beginning in the third period and accelerating a little bit more into the fourth quarter is simply, number one, due to our process, rescission benefit and rescission experience was trailing off. And then loan modification trends in those areas also began to trend down. And so, now we've taken that into our account into our reserve calculations. So we benefited from it back at the time when our experience demonstrated that we're getting that help, and we've had to correct, as we've seen that experience shift the other way as we've gotten through the lion share of some of these loan modification. So I guess that's how I would characterize what's going on there. You can see that's it's still a pretty significant percentage of our overall reserves are base in those areas, those across Michigan. A lot of the late stage stuff is really down there, but we continue to have cures coming out of there even at this point in time. And that's really a lot of the bases for the support what we have, and why we think we reserve at appropriate level. And lastly, when I think about what we've done, just to reiterate, we really went back and strengthened reserves on the oldest, later stage stuff. The new books aren't really feeding the beast anymore there. So the earlier performance on early term delinquencies are improving, so you're getting some benefit there. You're not adding more to the bucket. And then the stuff in the middle is although some of it continues, we saw some of the continued transition to foreclosure on the period, we're still getting some cures out of that.