Reviewing S&P/Case-Shiller

A cover story in the current issue of Big Builder magazine focuses on Dr. Robert Shiller, the Yale professor who helped found the Case-Shiller index in 1991 (and which partnered with S&P for marketing purposes a few years ago) and points out some of its flaws:

1. The index covers only 20 metro areas or 70% of the county (in California it misses some key areas such as the Inland Empire and the Central Valley), or less than what NAR covers.
2. The index weights higher-priced homes more than entry-level homes under the assumption that someone who is hedging the market is hit more by the upper-level sales (which may be true but is less useful for objective stats).
3. The index doesn't measure true value because it only tracks the costs of homes whose owners HAVE to sell for various reasons (and so ignores those who pull them off of the market).

According to the 5-year plan from TFS Derivatives Corp., which leverages the Case-Shiller index for housing futures, home prices in L.A. will shed 30% of their value between November of 2007 and 2012. Say what?

But I don't think it's that simple when applied to the micro market level. Almost all housing bears today simply pull the same string from their backs: "Housing prices rose much faster than incomes, so incomes will have to catch up -- either in real or nominal terms -- which means home prices will fall and then remain flat while incomes rise. Thank you and have a nice day, but please don't forget my speaking fee." Frankly, not an earth-shaking statement, and something most industry veterans were discussing in the early parts of the decade.

But what about rents? In a recent market study we did in the Hollywood area, we found that rents had risen by 51% over the last four years (and 72% since the year 2000) -- or also far above income growth. So are renters simply moving to cardboard boxes on the street or adapting by getting roommates, finding smaller apartments or relocating -- say it ain't so -- to the VALLEY?

On the landlord side, even if your home is currently worth less than it once was, why would you sell if you're a long-term investor (despite potentially better returns from T-bills) and can cover the monthly costs by renting it out? Wouldn't a certain floor value be maintained by average rents in an certain area and how large of a mortgage that those rents could support? So unless rents decline as much, I just don't think we're going to see prices decline to pre-2003 levels in all areas like some are predicting.

I think prices are going to change in specific neighborhoods not just relative to incomes, but also relative to (a) local rents; (b) how much equity local homeowners have; (c) proximity to employment centers and transit options; and (d) how many neighbors took on subprime, Option ARM and any type of stated income loans.

Consequently, to sweep your arm across Southern California and declare a 30% decline for every home in every neighborhood no matter the circumstance isn't just silly, but shows a troubling ignorance of how real estate markets really work. Could it happen in certain areas? Sure. But just like if the S&P loses 5% in a day, that doesn't necessarily mean the pain was felt evenly by every single stock in the index.