The housing market has softened much more than is being reported. We have been advising our retainer clients for more than one year about misleading national sales information, both with the existing-home sales and new-home sales data. We are now going public with our concerns because we are concerned that policy makers are relying on national data to conclude that the housing market correction has not been severe.
Here is our support:
Closing Data: We purchase and compile actual home closing data for approximately 181 counties across the country, which captures the counties where about 55 percent of the U.S. population lives and a significant percentage of all of the counties where the large home builders are active. This data shows that sales have fallen 22 percent if you compare sales over the last 12 months to the prior 12 months. On a straight year-over-year comparison, the decline is much more.
Mortgage Bankers Association (MBA) Data: The MBA seasonally adjusted purchase application index, which is a measure of the number of people filling out loan applications to buy a home, is down 18 percent from its peak in September 2005. With presumably more applications being filled out by borrowers who now have to shop around for a loan, how could sales have fallen by less than 18 percent?
Builder Data: The nation’s two largest home builders, D.R. Horton and Lennar, are reporting that orders have declined 27 percent to 37 percent year-over-year. D.R. Horton and Lennar have dropped prices significantly in many markets to generate sales, while the resale market has not. How could their sales have fallen more than the resale market, even if new-home communities tend to be in fringe areas?
Realogy Corp. Data: Realogy, which is the parent company of Century 21, Coldwell Banker and ERA, participated in roughly 1.9 million brokerage-related transactions in 2006 compared with 2.3 million in 2005, representing a year-over-year decline of 18 percent nationwide.
2005-2006 NAR State Data: The National Association of Realtors state data does show sharp year-over-year corrections in major states: 28 percent drop in Florida, 24 percent drop in California, and a 28 percent drop in Arizona. Our data, however, shows the sales have probably dropped by 34 percent, 27 percent and 38 percent, respectively. The national numbers include some large states where sales volumes have not corrected substantially, such as in Texas and Ohio, but we believe these markets are not very healthy for other reasons. Interestingly, our calculations were tracking very closely with NAR data through 2005, as illustrated above. We did investigate NAR methodology and have found absolutely no reason to believe that the NAR is intentionally misleading anyone, as some have suggested.
New-Home Data: The Census Bureau calculation of new-home data does not calculate sales net of cancellations, and cancellations are running much higher than normal right now, which is why the sales numbers overestimate actual sales.
The preponderance of evidence shows that the housing market in vibrant areas where home building is prevalent has corrected much more than some people believe it has.
In summary, we believe that the Fed should know that the housing market correction has been quite steep and is also not showing signs of bottoming out, as evidenced by all of the above information, as well as significant additional research we have conducted. While the Fed has far more to consider than housing, it should know that the housing market could sure use some lower interest rates to help achieve stability soon.
John Burns is the founder of Real Estate Consulting in Irvine, Calif., which monitors changes in real estate market conditions and provides consulting services, including strategic planning, market research and financial analysis. He can be reached at email@example.com.