Home > XE Currency Blog > The housing slowdown has arrived


XE Currency Blog

Topics6741 Posts6786
By New_Deal_democrat June 19, 2017 2:37 pm
  • XE Contributor
New_Deal_democrat's picture
New_Deal_democrat Posts: 547
The housing slowdown has arrived
On Friday housing permits and starts both declined, and significantly enough that they have become a negative.  Let's take a closer look.

First of all, here are permits for all structures (blue) and for single family homes (red):

I track single family permits because they are just as leading as permits overall, but much less noisy. They also avoid the distortion caused by a unique jump in NYC permits in May and June 2 years ago that has distorted the seasonally adjusted overall number.

What we see in both case was a surge in permits at the end of last year into early this year, and a significant downturn in the last few months.  In fact, this is the most significant downturn since 2010 (which itself was caused the expiration of a one year special program giving a big tax credit for homebuilding).

Starts are much more noisy, and tend to lag permits by a month or so:

Here too we see a leveling off at very least.

The best way to look at starts is via a 3-month rolling average which takes away most of the noise while only making a the series a little less leading. Unfortunately FRED doesn't have that capability, so let's look at the table supplied by the Census Bureau:

I'll let you do the  math in your head, but the bottom line is that the current 3-month rolling average of 1146 is the lowest in the last 12 months (in fact, a longer compilation shows it is the lowest in almost 18 months).

As I have noted many times before, the biggest single determinant of housing construction is interest rates. A secondary factor is demographics (the relative number of young adults of home-buying age). So let's take an updated look at a graph I have run many times: the YoY change in mortgage interest rates, inverted (blue) vs. the YoY% change in housing permits (red).  First, here's the long-term view:

And here is the close-up of the last 5 years:

YoY declines in long-term interest rates, which show up as positives in the graph, have been followed about 6 to 9 months later by relative YoY increases in new home building. Conversely, the big increase caused by the "taper tantrum" in 2013 was followed by a leveling-off in permits in 2014. In 2016 we first had record lows in interest rates (in July) which caused a spurt of activity at year end. Then we had the spike in interest rates after the November election. Accordingly I expected a slowdown by sometime this spring (see, e.g. here: http://community.xe.com/blog/xe-market-analysis/whither-housing-2017 and here: http://community.xe.com/blog/xe-market-analysis/housing-permits-and-star...) and it has clearly arrived.

In the past it has typically taken a decline of -200,000 permits from a peak to coincide with a recession. The least amount of -175,000 was in 1999-2000:

Friday's decline in permits was -132,000 off the January peak.

Let me make a few closing points. Housing permits and starts are only two among the number of long leading indicators. As I pointed out several weeks ago about the yield curve, reliance upon one single indicator is a recipe for getting it wrong. 

Further, that housing has turned negative now doesn't mean it will continue as a negative in months to come. The biggest mistake most analysts make is in simply projecting the current trend forward.  That being said, what looks most likely is for housing permits to remain a neutral to negative in the next few months, as YoY interest rates are currently still higher. Interest rates say negative, but housing still has a demographic wind at its back.

And on interest rates themselves, much will depend upon the Federal Reserve. Long term rates have reacted to the last two moves by declining somewhat, as the bond market foresees a weaker economy. Will the Fed hike yet again, risking an inverted yield curve? Will mortgage rates fall all the way to new lows, re-igniting demand?  That is the subject of another post in the future.

For now, the new home sales report on Friday assumes added importance. It is a very noisy and heavily revised metric, but it does tend to peak if anything before permits and starts. And it just made a peak three months ago.  Will it hold up, or follow permits and starts down? We'll see on Friday.

Sent from my iPad

Paste link in email or IM