Timothy Wadhams
Management
Sure. Basically, Steve, that is an annual test that we do that is performed in the fourth quarter unless there is a trigger - what's called a triggering event - that might cause you to look at one of your business segments earlier than that. So we do an annual test every year in the fourth quarter and that test is based on a five-year discounted cash flow model. There was a fair amount of disclosure in the 10-K this year in terms of methodology and assumptions and basically one of the key issues there is really the terminal value when you do that calculation. When you get out into the fifth year - and as I think I mentioned on the fourth quarter call - the assumption that we had made when we did that calculation last year was that housing starts were going to be back at about 1.5 million - 1.55 million, I believe. And, again, that's in the 10-K. So what we've told investors is that as we start to put that analysis together again this year in the fourth quarter, the key drivers of that are basically going to be assumptions as to what we think housing starts will look like over the five-year period, with that fifth year being very important in that calculation. So at this point in time, again, we haven't really performed that exercise at this point and from our position right now it would be pretty hard for us not to see a start number that probably is going to be around 1.4, 1.5 five years out, which we think is reasonable. Last year we didn't pull that number out of the air. We basically went to, I think, Global Insight and some other consensus opinions, if you will, from economists, took their number, discounted that. So from that standpoint we continue to think that we'll probably have a number in that range. The one thing I would mention to you is that if you think about the longer term, you think about demographics, you think about household formations, we continue to think that the fundamentals are favorable. I think the Harvard Joint Center for Housing Studies recently estimated that household formations in the decade 2010 to 2020 would approximate, I think, 14 million, if my memory is right, and maybe with an average of 1.6 or 1.7 per year - obviously, a big driver in terms of new home construction. So from our standpoint at this stage, again, we've got to get into the detail but I wouldn't necessarily anticipate that we're going to have a major issue based on what I think those projections will look like when we get to the fourth quarter.