Case-Shiller Home Price Indices data through November 2009 reveal the annual rates of decline of the 10-City and 20-City Composites are improving. There were price declines measured across many markets during November.

This was the 10 month of improved readings — less bad price declines — in the annual statistics. It was also the third consecutive month of “only” single digit drops, following 20 consecutive months of double digit price declines.


Source: Standard & Poor’s and Fiserv

The 10-City and 20-City Composite (above) declined 4.5% and 5.3% respectively.

Source: Standard & Poor’s and Fiserv

Category: Real Estate

Please use the comments to demonstrate your own ignorance, unfamiliarity with empirical data and lack of respect for scientific knowledge. Be sure to create straw men and argue against things I have neither said nor implied. If you could repeat previously discredited memes or steer the conversation into irrelevant, off topic discussions, it would be appreciated. Lastly, kindly forgo all civility in your discourse . . . you are, after all, anonymous.

10 Responses to “Case Shiller’s “Mixed Messages in the Data””

  1. wunsacon says:

    The only time home prices remained stable for some years was during the period from 1989 to1997. Is “75″ a more appropriate baseline than the “100″ printed in 2000? During the height of a bubble?

  2. DeDude says:

    The Y-o-Y % change is the least informative graph you can use. It is very late in demonstrating shifts in a parameter. I am a little surprised to see a second peak developing in the index number itself. Wonder if that is an effect of prices raising in the low end because that end of the market is where all the action is (so it is getting a little crowded at that end of the price range).

  3. rootless_cosmopolitan says:

    FYI: The Case-Shiller index is constructed from a three-month moving average of home prices. The index value of a month is derived from home prices of this months and the two preceding months. See:–p-us—-

    and then look in the pdf-file with “Methodology” to be found as one of the menu items on the left side, page 6 under “Approaches”.

    Therefore, one can’t draw any conclusions about home prices in a particular month compared to any other month from looking at the index level and its change. Prices may very well already go into the opposite direction compared to what the index says due to the three-month averaging. Insofar, even S&P’s press release talks nonsense where it makes a statement not just on the index level, but on home prices for November compared to October 2009 or November 2008.


  4. jpm says:

    The Y-o-Y % change is the least informative graph you can use.

    Not when you consider that prices are seasonal with a period of one year.

  5. Marcus Aurelius says:

    Charts are good, and all, but the basic laws of supply and demand don’t bode well for the future of housing in causing or reflecting an improvement in the economy. When we can honestly say we’re improving is when mortgage/income and rent/mortgage ratios return to normal levels, all shadow inventory is accounted for and marked to market, all underwater MBSs are honestly valued and unwound, and unemployment levels are unremarkable for a healthy economy.

    Like I said — other than that, nice chart.

  6. DeDude says:

    @jpm, as I understand it the issue of seasonality os already baked into it – its an index. That is why it doesn’t have summer spikes and vinter slumps.

  7. jpm says:

    as I understand it the issue of seasonality os already baked into it – its an index.

    Seasonality is not baked into it.

    That is why it doesn’t have summer spikes and winter slumps.

    When plotted as something other than YOY, it definitely has summer spikes and winter slumps. For the raw and SA data, see CRs plot:

    Compare CRs plot to the YOY above, and you see why some folks examine YOY for Case-Shiller. (BTW, I’m not claiming the plot above is the be-all-and-end-all, but it is certainly a useful piece of info.)

  8. mknowles says:

    I wonder how that looks juxtaposed with the upcoming alt-a/option arm wave, especially in California. I expect the banks will drag that out as long as possible, to prop up prices.

  9. DeDude says:

    jpm; yes Calculated risk (a great fact based blog that is on Berry’s Blog Roll) has a new entry about the seasonal adjustment of the Case-Shiller index. They are comparing price data that is or is not seasonally adjusted and show that without seasonal adjustment you get a shark teach pattern. They claim the Case-Shiller index is doing a pretty good job of getting rid of seasonality (i.e. compare shark teach data with Case-Shiller), although they also claim that it is possible that there is a little more trouble lately with that (not sure what that claim is based on). My biggest problem with seasonally adjusted C-S is that I find it visually much more difficult to pick up on the shoulders of a graph (the stop-in-the-drop). It does give you information but it is a lot more difficult to pick up changes in a trend when they occur.

  10. DeDude says:

    Said a little differently; the Y-o-Y change numbers can change because something drastically happened last year or because something drastically is happening now. The actual numbers will not have that confusion (although if not seasonally adjusted they can have seasonal confusion). The best way to deal with both issues is the way Calculated Risk (often linked to by Barry) show sales numbers for housing (on a 12 month graph with different colors for each year). That way you can actually sort out seasonality in an easy visual way. Wonder how Case-Shiller numbers would look in that format ;-)