The divergence between the NAHB index and other housing indicators has continued to suggest that sentiment was “getting ahead of itself" and as Citi's Tom Fitzpatrick warns would suggest that the qualitative nature of the overall housing recovery is less robust than one would like. Housing should pause/consolidate possibly even for most of this year as the weather argument that is trotted out by so many commentators does not seem to hold up to even a basic examination with the worst data coming from the West Coast. Simply put, Citi warns, we think housing sentiment got carried away as it did into 1994 and 1998 post the housing/savings and loan crisis of 1989-1991.
Via Citi FX Technicals
Techamentals - Housing Data Doing What Was Expected
The surge higher in yields seen last year along with elevated oil prices, some EM stress, and Fiscal drag seem to finally having their effect.
The weather argument that is trotted out by so many commentators does not seem to hold up to even a basic examination with the worst data coming from the West Coast (Maybe that 70 degree heat was just too much for them)
The chart below has constantly argued that housing should pause/consolidate, possibly even for most of this year before likely thereafter showing some traction again.
NAHB index; Building permits; New home sales; Housing starts
As we saw after the housing/savings and loan crisis in 1989-1991 housing sentiment (NAHB) rose much quicker that actual activity (Permits, New Home Sales and Housing starts)
We saw this in 2 periods in particular
- Into late November 1993 before the 13 month surge in 10 year yields (287 basis points) as Alan Greenspan “tinkered” by raising interest rates by 25 basis points in Feb. 1994.
- Into December 1998 before the EM crisis, Russian default, LTCM failure also became a drag on sentiment.
Once sentiment corrected and converged to “reality” we saw housing recover again. That took about 12-18 months in the periods in question.
We would therefore not be at all surprised if housing continues to consolidate/correct for most of 2014 before once again resuming its gradual rise higher again.
What would a path like 1993-1995 suggest?
A move lower in the NAHB index into the start of the 4th quarter to a level around 27 before resuming its rise would fit with what we saw in 1993-1995.
Mortgage Bankers association purchase index
This shows mortgage loan applications submitted to lenders and shows that we are almost back to the low levels seen in August 2011
The peak here (not surprisingly) was seen in April 2013 before we saw a surge in mortgage rates between May and July
In early May 2013, 30 year mortgage rates were around 3.40% and then surged by July to 4.64% (36% higher). Today they still remain elevated to last year (Albeit off the highs) at 4.32%
This area between 156.80 and 159.30 above (76.4% pullback of the 1990-2004 rise and horizontal supports from 1996 and 2011) is big support. If that were to give way then a move back towards the lows set in late 1990 at 53.50 would be a danger.
What this really shows is that a lot of housing activity has not been the traditional taking out of mortgages to buy a home but rather cash purchases; buy to rent; distressed purchases etc. This would suggest that the qualitative nature of the overall housing recovery is less robust than one would like.
Housing affordability: Small bounce but still 18% off the levels seen in March 2013
Rising mortgage rates being the prime culprit of course
So to improve affordability and get housing moving again we are going to need to see lower mortgage rates. How does that look?
US 10 year yield weekly chart: Lower yields in the months ahead remains our base case
Last month we saw an outside month in the 10 year yield (not shown) that suggests lower yields can be seen.(Traded to the high of the trend at 3.05% then fell below the December 2012 low of 2.75% and closed in January below that level at 2.64%). We also saw an outside month on the 30 year yield with the close below 3.74%
Going back to the chart above we look to have a clear Double top potential with a neckline at 2.47%. A close below there if seen would suggest the potential to go as low as 1.90-1.95% again. In addition to the neckline we see significant horizontal trend line support as well as the 55 and 200 week moving averages converging in the 2.40-2.44% range. So bottom line we continue to expect a test of this support area at 2.39-2.47% at a minimum. A weekly close below here, if seen, would suggest extended losses that could take us below 2%.
On the 30 year yield chart the picture is similar with significant support in the 3.48-3.56% area. A weekly close below there would suggest extended losses towards 3.15-3.20%
Overlay of US 10 year yield chart and 30 year mortgage rate.
A move towards the initial support area around 2.40-2.47% on 10 year yields would suggest a drop in the mortgage rate to at least 4.05-4.15% from the present 4.31% while a completion of the double top and a move below 2% would suggest mortgage rates back towards 3.65% at least.
It is feasible the move in mortgages could be even more if the spread were also to narrow as we moved lower.
US 30 year mortgage rate minus 10 year yield
Testing rising trend line support coming from 2007 around 152 basis points. A break below here would suggest further narrowing in this spread which if coming in a downward moving yield environment would go some way towards improving housing affordability again
Summary: We think housing sentiment got carried away as it did into 1994 and 1998 post the housing/savings and loan crisis of 1989-1991.
The surge in yields since last May was “too far too fast”. Add to that the fiscal drag, elevated oil prices and maybe even the weather (as A factor not THE factor) and you get a pause in housing and a fall in sentiment like we did in 1994 and 1998. With sluggish economic data materializing, yields and ultimately mortgage rates will adjust lower (without the need for additional Fed interference) as the bond market “clears” all on its own. This will be simulative and by as early as end 2014 housing will likely pick up once again.