Update: Added quotes from Robert Shiller.Like most measures of the economy, the S&P/Case-Shiller home price index is not perfect. However, it has a critical shortcoming that almost no one talks about.
We already know that the data comes on a bit of a lag. Today’s Case-Shiller numbers reflect home prices that were recorded back in November.
But are they really the November home prices?
The short answer is a resounding NO.
Business Insider spoke to Kevin Caron, strategist for Stifel Nicolaus:
The data doesn’t hit the database until the public filing after closing. But the closing may be months after the agreement between buyer and seller (and the banks that provide financing). Ultimately, the lag can be a long time (maybe six months) between when a price is agreed upon, the mortgage is secured, the closing occurs, and the sale is recorded and available for public use.
David Blitzer, Managing Director and Chairman of S&P’s Index Committee, told Business Insider that there is indeed a lag between the time a price is agreed on and when the sale closes. However, he thinks a six month lag may be on the high side. We asked Blitzer about Caron’s assessment. Here’s how he responded:
The analyst is correct about one point — prices for S&P/Case-Shiller (SPCS) are based on public filings. However, these are the only consistent, reliable and accurate source of price data. Further the time from contract to closing is more like 4 to 8 weeks, not 3 to 6 months. Also, many sales collapse before the sale closes and the only way to be sure the sales is arms length is through the public records of closings.
We also spoke to Professor Robert Shiller, co-creator of the Case-Shiller index, who shared Blitzer’s sentiment:
Our indices are based on closings. But I think that is the right thing to do. Many purchase and sales agreements are never consummated. Those that are away from the market are less likely to be completed. Using purchase and sales agreements to construct an index would be like taking limit orders on the stock exchange to compute a stock price index as if they were transaction prices.
Blitzer and Shiller both make good points about practicality. Like Blitzer said, “these are the only consistent, reliable and accurate source of price data.”
We won’t argue that the Case-Shiller data is certainly robust in terms of its consistency, breadth and depth. Over long periods, useful trends reveal themselves in the Case-Shiller data.
However, four to eight weeks from contract to closing is major lag. Based on Blitzer’s estimate, today’s November home price data reflects September or October prices at contract, which is the more relevant measure for a home buyer or seller. In other words, it would be inaccurate for users of the Case-Shiller data to assume that the monthly index data reflects monthly market prices without some additional lag.
Furthermore, the time from contract to closing may vary depending on the city, which would make the Case-Shiller indices even more problematic.
Sure, no measure of the economy is perfect. GDP growth and unemployment rates are revised frequently. And no one will forget when the National Association of Realtors (NAR) unleashed the mother of all revisions when it slashed four years of existing home sales data by 14%.
However, those using any data should be aware of what the data is really saying.