Tomorrow morning's existing-home sales report for June from the NAR (National Association of Realtors) is widely expected to receive a boost from the expiration of the federal home purchase tax credit at the end of that month. However, analysis of pending home sales data (also from the NAR) leads to a projection for a low number of existing home sales for June.
The NAR describes the value of pending home sales data as follows:
Specifically, Pending Home Sales become Existing-Home Sales one-to-two months later. This means that we can use an index derived from Pending Home Sales to predict actual home sales activity.Since pending home sales measure actual existing-home sales, the PHSI provides an accurate and reliable indicator of future home sales activity. Our sample shows that over 80% of all pending home sales go to settlement within a 2-month time-period (and a significant share of the rest close in month 3 and month 4).
Because of the expiration of the home purchase tax credit, some closings may have been accelerated into June. That occurrence could introduce a bias into the historical correlations. Even so, the projections from those correlations are sufficiently low that such bias could still leave existing home sales below what many are expecting.
The only significant correlation that this analyst finds in recent years is between pending home sales and existing-home sales one month later. The correlations two months or more later are near zero or negative. However, that correlation is weak (correlation coefficient = 0.61). This has prompted further analysis.
Both existing home sales and pending sales data is quite noisy. Because of this, the effect of using two-month moving averages to smooth the noise has been examined. The correlation coefficient is much improved to 0.84. The result of a correlation plot for data from January 2008 through May 2009 is shown in the graph on the next page.
The solid blue line is the linear trend line for the 29 data points. The dashed blue lines bracket 50% of the data and the dotted blue lines bracket all 29 data points.
Projecting Future Existing-Home Sales
Projections are made for the most likely June existing-home, two-month moving average when the June data is announced on July 22 (trend line value labeled as "mid-range," even though it is slightly off center). The upper and lower bounds that include the range of 50% of the data points and the upper and lower bounds that include all 29 existing data points (labeled 95% probability limits) are also shown on the graph. The value of probability of 95% is an estimate, considered conservative at this point in model development because, if the next data point is outside that limit, it would be one point out of 30. That means 96.6667% of data would be within the limits, a number larger than 95%.
The June existing-home sales number can be projected from the graph above. We start with the value of the two month moving average for pending home sales as of the July 1 data, which is -12.0%. We also have the single month existing-home sales number for May (June 22 data), which is -2.2%. The trend line indicates that June existing-homes sales will be down by 10.85%, which is calculated as follows:
2-Month M.A. Existing-Home Sales = y = 0.9057 x - 0.0012
= 0.9057 * ( -0.1200) - 0.0012
= - 0.1085 = - 10.85%
Since May is -2.2%, the projected June value is - 19.5%.
The projection for tomorrow's existing home sales is 4.56 million, annual rate. With the uncertainty brackets, the projection is as follows:
With 50% confidence, June existing home sales = 4.56 +/- 0.18 million.
With 95% confidence, June existing home sales = 4.56 +/- 0.24 million.
These projections are for June existing-home sales to be no greater than 4.80 million, with 95% confidence. This is far below the May reading of 5.66 million and below any other estimates I have seen.
Caveat: Existing-home sales may be moved forward from July (increasing the June number) because the tax credit requires contracts signed before April 30 and closing to occur by June 30. If that were to occur it will just drive the July existing-home sales number lower than it would have been otherwise.
If this projection is realized, home builders will likely have a bad day tomorrow. In spite of the declines for homebuilder stocks since early May, there may be a bad reaction in these stocks. Examples of stocks to watch include
. All three are 28% to 31% below highs on May 3, but have rallied 6% to 9% since early this month. The homebuilder ETF
could also have a bad day.
Disclosure: No positions in stocks mentioned.
John B. Lounsbury is a financial planner and investment adviser, providing comprehensive financial planning and investment advisory services to a select group of families on a fee-only basis. He worked for 34 years with IBM, and spent 25 years in R&D management and corporate staff positions. He also was a Series 6, 7, 63 licensed representative with a major insurance company brokerage for nine years.
Specific interests include political and economic history and investment strategy analysis. He holds degrees from the University of Vermont, Columbia University and the Illinois Institute of Technology, where he studied chemistry, physics and mathematics. He is a contributor to Seeking Alpha and his own blog,