Substantial Future Home Price Declines Predicted By Goldman Sachs And Peak Theories
For anyone following the recent collapse in mortgage applications, the recent "strength" in new and existing home sales is nothing but the latest joke to spin the nth bounce from the bottom as the "this is it" moment which Cramer has been trying to do with disastrous results ever since the summer of 2009. Oddly, reading a recent surprisingly bearish Goldman economic outlook (or not so surprising: it lays out the framework for Goldman to start advocating MBS purchases as part of QE3) piece from Sven Jari Stehn confirms our concerns that any attempt at shining light behind the headlines exposes ever more cockroaches. In "Mortgage Applications Point to Near-Term Home Sales Weakness" Stehn highlights the same issues we have been pounding on the table for months: namely that near contemporaneous plunge in mortgage applications is far more troubling and should be given far more impact than new, pending and existing home sales in any one prior period. Goldman summarizes: "The number of mortgage applications, however, has declined sharply in
recent weeks. Specifically, the volume of mortgage applications for
purchase—reported in a timely fashion every week by the Mortgage Bankers
Association—declined by a cumulative 14% during the last three weeks.
Does the decline in mortgage applications suggest that home sales are
set to decline again in coming months?" In short the answer is yes, and the full note below explains it. Additionally, we have provided some technical perspectives from Peak Theories which predict a 7% drop based on recent chart patterns. Needless to say, we believe the drop will be far greater when all is said and done, now that the Bernank has given up on attempting to keep mortgage rates low and only cares about boosting stock prices.
Mortgage Applications Point to Near-Term Home Sales Weakness
Following consistently disappointing news out of the housing market since the summer, indicators of housing activity have recently looked somewhat better. Existing home sales rose 6.8% and 11.8% in November and December, respectively. Moreover, new home sales surged 17.5% in December. (Note, however, that this increase was due almost entirely to surging sales in the West of the United States—possibly driven by an expiring tax credit in California.) Meanwhile, the rate of decline in home prices appears to be stabilizing. In particular, the Case-Shiller index declined at a slower rate in November and, as the index is a three-month moving average, home prices might have stabilized in November on a sequential basis.
The number of mortgage applications, however, has declined sharply in recent weeks. Specifically, the volume of mortgage applications for purchase—reported in a timely fashion every week by the Mortgage Bankers Association—declined by a cumulative 14% during the last three weeks. Does the decline in mortgage applications suggest that home sales are set to decline again in coming months?
The chart below plots the month-to-month change in existing home sales alongside the (monthly average of the) volume of mortgage applications since June 2009. The number of mortgage applications indeed appears to lead existing home sales. For example, mortgage applications dropped sharply in May 2010, before existing home sales declined in June and July. Mortgage applications then picked up during the summer, again leading existing home sales. Given the recent drop in mortgage applications, this relationship would imply a substantial drop in the growth rate of existing home sales during January and February.
A set of simple regressions supports this finding. Specifically, we estimate models that explain the growth rate of home sales with the change in the number of mortgage applications. We estimate three separate regression for existing, new and pending home sales between January 2009 and December 2010 (see table below). We make two observations:
First, the regressions confirm that the volume of mortgage applications holds useful information for home sales. In particular, we find that existing home sales depend on the number of mortgage applications with a one month lag. New and pending home sales, however, are most closely associated with the contemporaneous number of mortgage applications. This is not surprising given that sales of existing homes are measured at closing and the application for a loan typically precedes the closing by several weeks. New home sales, in contrast, are measured based on the signing of a sales contract, at which point the buyer still has time to apply for the mortgage.
Considering a longer sample—back to 1990 for existing and new home sales—leaves our finding qualitatively unchanged but reduces the fit of the regressions and the size of the estimated coefficients. (Note, however, that care has to be taken in including the period between late 2006 and late 2008. During this time the link between home sales and mortgage applications weakened substantially, reflecting upheaval in the mortgage finance industry. The rising difficulty of obtaining credit seems to have resulted in a higher number of applications per loan granted.)
Second, the recent decline in the volume of mortgage applications points to a decline in home sales in the near term. Given the one-month lag between existing home sales and mortgage applications, the model suggests that existing home sales will slow to a growth rate of about 2% in January and then decline by around 4% in February. New and pending home sales are predicted to decline by around 4% and 5% in January, respectively.
Taken together, the predictions from the recent decline in mortgage applications are consistent with our view that the housing market will remain weak in 2011. In particular, we expect only a moderate pickup in housing starts and home sales throughout 2011.
Follows the latest report from Peak Theories Research: Housing Prices to Decline
- advertisements -