Housing Data Is Old and Moldy

Data to gauge the housing market is hobbled by methodological problems, many of them related to seasonality and the lag from the time a contract is signed to when a sale is recorded.
A trailing indicator.

Consumers rely on a monthly parade of reports to get a sense of the state of the U.S. housing market, thinking that the data offer near-instantaneous insights in much the way that a stock-market index does.

Seeing housing this way is a mistake: The information available is hobbled by the very nature of the market itself. Buying or selling a house is a slow, lumbering, inefficient process that takes months to complete. There is a significant lag time from a meeting of the minds between a buyer and a seller to when the sale gets reported to the public. Yet we hang on each report and the bold headlines they inspire without proper context or understanding.

By taking a look at the two housing reports released this week we can see why the information available to the public is so deficient, if not misleading.

The most notable was Tuesday's S&P/Case Shiller report -- actually, a series of reports. However high your regard may be of Case Shiller and its co-creator, Yale economist and Nobel laureate Robert Shiller, there are real weaknesses in the methodology.

On the surface, the latest 20-City Home Price Index seemed to suggest some slowing in the market, with a 9.3 percent rise in existing housing prices in May, a slower rate of increase than in the preceding two months.

This edition of the S&P/Case Shiller report, although labeled "May," is the moving average of completed sales from March through May. If we assume their respective sales contracts were signed 60 days before closing, the report reflects the market period from January through March. And because it often takes several weeks from the time a buyer and a seller agree to terms and sign a contract, 30 to 60 more days might pass. This means the index probably reflects deals reached in the period from December through February.

If we pick January as the midpoint of the range, then the latest Case Shiller report represents what buyers and sellers were thinking in the beginning of the 2014, just about the time the Northeast and Midwest were enveloped in the brutal cold of the so-called polar vortex. The inclement weather is blamed for the 2.1 percent decline in first-quarter economic growth. To make this index even less meaningful, it measures housing prices, the least current metric of all. Housing prices are the end result of the interplay of sales and inventory trends.

The other U.S. housing market report was Monday's Pending Home Sale Index, or PHSI, released by the National Association of Realtors. The NAR is an industry trade group and dependable cheerleader for the residential housing market. In spite of this, its index has a 30-day lag time, the shortest of the various real-estate market reports. It also has the advantage of reporting the number of contracts signed on existing homes, as opposed to prices.

The NAR has long described the index as a "forward-looking indicator." In reality, the PHSI is simply less backward-looking than any other national housing index. For that reason alone, I consider it among the best gauges of the U.S. housing market, despite my reservations about the credibility of NAR's always-sunny commentary about residential real estate.

The latest report showed that signed contracts in June fell 1.1 percent from May and were 7.3 percent lower than a year earlier. Because the housing market is seasonal in nature -- sales peak in spring and fall -- these numbers are adjusted to account for this normal volatility.

But look at the nonseasonally adjusted results: The change from May to June was a 0.6 percent increase and the year-over-year decrease was 4.5 percent. Here is how you need to think about these numbers. The first half of 2013 saw an acceleration of pending home sales, the product of five years of pent-up demand. This resulted in a surge in contract signings, making the year-over-year comparison look weaker than it really was.

The housing market is recovering, but still in a slow grind as we wait for gains in income, job growth and credit availability. Until that happens we will continue to get confusing messages from research that may capture the past but doesn't tell us enough about the present or the future.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

    To contact the author on this story:
    Jonathan J Miller at jmiller@millersamuel.com

    To contact the editors on this story:
    James Greiff at jgreiff@bloomberg.net
    James Greiff at jgreiff@bloomberg.net

    Before it's here, it's on the Bloomberg Terminal.