The Morning Report
San Diego news and info
you need to take on the day.
The median price data had indicated some home price stabilization through March, but the latest release of the Case-Shiller index suggests otherwise. I’ll talk a bit more below about the strengths and weaknesses of the CS index, but first let’s have a look at the latest data.
The media has made much of the fact that the national Case-Shiller index hit a new post-boom low, officially ushering in a “double dip” in nationwide home price. But while San Diego prices did drop in March, the CS index indicates that we still held up above those 2009 lows:
Zooming out to the entire post-boom period starting in late 2005, however, it’s clear that we aren’t substantially above the lows in the grand scheme of things:
If we adjust for inflation as measured by the Consumer Price Index, the high-priced and mid-priced tiers are actually now below their prior lows. The low-priced tier and the aggregate index are still above the 2009 inflation-adjusted lows — though just barely, in the case of the aggregate index. So while San Diego has avoided the nominal double dip, from an inflation-adjusted standpoint we are nearly there.
Here’s a look all the back to 1989, also adjusted for inflation:
As I mentioned earlier, the March Case-Shiller index registered a price decline while the median price per square foot indicated stable prices for that month. Of these two measures, the CS index is better — but it definitely has some flaws of its own.
The huge advantage in the CS index is that is calculated based on same-home sales. The median price tells us how much the typical buyer paid, but says nothing about what that buyer got for the money. Using a median price per square foot helps a little by telling us how big a place the typical buyer got for the money, but it says nothing else about the quality of the home or its neighborhood. The CS index addresses this issue by tracking changes in the price of individual homes and aggregating those changes together to get an idea of market-wide price changes.
But while this is a major step in the right direction, the CS index still has some issues. For instance, what if the condition of the home changed due to renovation or deterioration between its first and second sales? How do we know how much of the price change was caused by a change in the condition of the home as opposed to changes in general market pricing?
The answer is that there is no way to know. The CS calculation methodology attempts to mitigate this issue in a couple of ways. First, the “weight” of a home is reduced — that is, it is given less of an influence on the calculation than other homes — if its price change is significantly different from other homes that sold during the same period. But that wouldn’t necessarily help in a situation where homes were generally being over-improved (such as during the bubble) or are under-maintained (such as now, when foreclosures in poor condition account for an unusually large number of sales) on a market-wide basis. A second adjustment that the CS calculation makes is to lower the weight of homes that went a long time between sales, because it’s more likely that the condition of such homes changed due to the longer timespan involved. This surely helps with the home-condition issue, but it could introduce its own distortion in that older areas of San Diego (where the average period between sales is longer) may have less of a weight in the index than newer areas, thus giving newer areas more influence on the region’s calculated price changes. (Thanks to commenter FryeFan for pointing that one out).
There are other potential issues as well. I’ll spare you the details (gluttons for punishment can read a previous article with some further thoughts on the subject). The main point is that the calculation of the Case-Shiller index involves a good amount of statistical massaging to arrive at that countywide monthly number.
And there are other issues which aren’t specific to the CS index but would impact any attempt to sum up the whole market in a single number. A single estimate of price changes across multiple sub-markets will always be distorted by the fact that the many different areas of San Diego could be (and are) experiencing significantly different market conditions from one another. And since price changes can only be calculated based on what’s actually selling, a segment of the market that is selling unusually fast could have an exaggerated influence on the market-wide price calculation.
So the Case-Shiller index is the best price indicator we have, but it is far from perfect — and in fact, no single countywide price indicator could possibly be perfect. This is something to keep in mind as we watch these numbers from month to month.