By Dirk van Dijk
Yesterday we got news that New Home Sales had fallen to their second all-time record low (since the data started in 1963) in a row. With other areas of the economy clearly on the mend (at least for now) the question arises: when will housing come back?
I want to focus on new home sales, not existing home sales, because new home sales directly contribute to economic growth, while existing home sales have only an indirect effect. The U.S. is a much larger place now than it was back in 1963, both in terms of population and the overall size of the economy, even adjusted for inflation. With more and richer people, one would expect that there would be a demand for more and newer houses.
In the graph below I show both population and new home sales, both set to an index starting in 1963 (when the new home sales data starts). Notice that over time, the area over the blue population line roughly matches the area under the line, or at least it did until the housing bubble took off after it really started to inflate around the end of the 2001 recession.
Reading the Numbers
While the math of figuring out the area above and below the curve gets a little complicated, visually it is pretty easy to see. Given the huge area above the line built up from about 1996 through 2007, it is not surprising that we have fallen well below it.
The area under the line has been accumulating rapidly over the last two years. If we were to immediately return to the line (not going to happen, but just a thought exercise), the area under the line would still be less than the area above the line. That, however, would not preclude a recovery over the next few years.
Just from visual inspection, it looks like if we returned to the line by mid-2013, we would have more or less worked off the excess housing that was built up during the housing bubble. Given the very large inventories of both new and used houses on the market right now (including the shadow inventory of existing homes where the owners are either very delinquent on their mortgages or are already in the process of foreclosure), it is very clear that we are not at the point yet.
There are currently 9.2 months of supply of new houses on the market at the current sales rate, and 8.6 months of supply of existing homes. For both of those, a healthy market would be between 5 and 6 months (both were much lower during the housing bubble, around 4 months or so on average).
However, just returning to the population line would result in a near tripling of the level of new home sales. That would be a compound annual growth rate of over 40% if it happened over the course of three years, which would also be roughly the time needed for the area under the line to equal the area over the line that was accumulated between the mid 1990’s and when sales broke below the line in early 2007.
Part of the problem right now is that even though the population is growing, the rate of household formation is very low. Household formation is basically economist-speak for people striking out on their own. For young people in their early 20s, getting out of Mom and Dad’s house and getting a place of their own. For them to do that they need a good steady job.
As we start to increase employment, the excess housing inventory, both new and used, will start to get absorbed. Unfortunately, historically construction jobs have been a major swing factor in employment. If there are no new construction jobs, the overall rate of job creation will be much lower than it would be otherwise.
Commercial real estate is also very overbuilt, and has very high vacancy rates in almost all major areas (both geographically and by type, i.e. office, retail etc). Thus we are not likely to see a lot of construction jobs being created on that side anytime soon.
Recovery & the Construction Dilemma
While construction employment peaked sooner than did overall employment (Aug 2006 at 7.725 million), it was still relatively strong when overall private sector employment peaked in November 2007 (7.539 million). At the peak, there were a total of 115.551 million private sector jobs, and in February there were 107.056 million jobs, for a loss of 8.495 million jobs. In February, there were 5.555 million construction jobs, a loss of 1.984 from when the recession started and of 2.170 million from its own earlier peak.
In other words, when the recession started, construction was responsible for 5.19% of all private sector jobs in the country, yet that sector has accounted for 23.4% of all private sector jobs since the recession started.
We thus have a bit of a "chicken and egg" problem. Demand for housing should pick up when employment picks up, and employment should pick up when demand for housing picks up. If we can get the cycle flowing, it should be self-reinforcing, but how to break the cycle is a bit of a conundrum. Construction is also noteworthy in that it tends to pay relatively good wages to relatively poorly educated (at least formally educated) people.
Historically, residential investment, of which new home construction is the most important part, has been the locomotive that has pulled the economy out of recessions, a quick glance at the relationship between the red line and the grey recession bars will confirm that. This time, residential investment looks like at best it will be one of the box cars that is dragged along by another engine. What that engine will be is a bit of a mystery, although demand for other durable goods does seem to have picked up a bit, so that might be part of the answer.
Housing Could Be a Powerful Force
However, later on, once the down cycle is broken and the self-reinforcing virtuous cycle gets underway, housing could be a powerful force driving the economy higher. One does not have to make very heroic assumptions about the absolute level of future new home sales to generate very impressive percentage gains -- just set the base low enough -- and we certainly seemed to have accomplished that.
As housing recovers, not only will the homebuilders like Lennar (LEN) and D.R. Horton (DHI) benefit (right now they are being kept on life support due to tax loss carrybacks that mean just about every penny of taxes they paid during the boom is coming back to them), but all the firms that supply the homebuilders will also gain.
That means a huge part of the economy, ranging from the heating and air conditioning division of United Technologies (UTX) to lumber companies like Plum Creek Timber (PCL) to plumbing suppliers like Masco (MAS). Of course, if those construction workers go back to work they will be able to go out to stores, ranging from Big Lots (BIG) to Nordstrom (JWN) (OK, maybe more Big Lots than Nordstrom).
The turn in housing will come, but probably not right away. But when it happens, it will be powerful, at least on a percentage basis, and that will mean lots of good things for lots of different areas of the economy.