The party in Housing Starts and Building Permits is over, and the industry is now dealing with the hangover. With the homebuyer tax credit now over (people have until the end of June to close, but contracts had to be signed by the end of April), housing starts have plummeted.
In May, total housing starts fell by 10.0% to a seasonally adjusted annual rate of 593,000. That is up just 7.8% from the extremely depressed levels of a year ago. The level of starts was well below consensus expectations of a 655,000 annual rate.
Even those dismal numbers understate the damage. For starters, April’s numbers were revised down to an annual rate of 659,000 from an originally reported 672,000. Thus, from where we thought we were, the drop is more like 11.8%.
Further, the number was cushioned by a big increase in the very volatile apartment and condo segment. The number of starts in buildings of five or more units soared 38.3% to an annual rate of 112,000, although that is still 17.0% below the year-ago level.
When most people think of housing starts they think of single-family houses. Those declined 17.2% on the month but are up 15.3% from a year ago. The graph below (from http://www.calculatedriskblog.com/) shows the history of housing starts, both total and single family.
The damage was concentrated regionally in the South, which is by far the largest, and thus most important of the four census regions when it comes to housing data. Housing starts in Dixie plunged by 21.3% to an annual rate of 228,000, although that is up 4.7% from a year ago. In other words the region was responsible for 38.4% of all starts, down from 55.5% in April.
The April percentage was about normal; the May percentage was the lowest seen in a very long time. The Northeast, which is the smallest of the four regions, was the next weakest with a 6.3% decline, but up 25.0% from a year ago. The other two regions actually saw an increase in housing starts, with the Midwest seeing a 4.9% increase on the month and up 35.4% year over year, while the West saw a 10.8% jump on the month, but it is still down 9.6% year over year.
Housing Starts, Historically
As the graph shows, normally housing starts begin to fall before a recession starts, but then rebound strongly just before the end of the recession. Residential investment is thus normally one of the most important locomotives in pulling the U.S. economy out of a slump (and declines in housing starts one of the key reasons we go into recessions). That locomotive has been derailed so far in this recovery, and that is one of the most important reasons why this recovery has been so anemic.
The 2001 downturn was an exception to the rule about housing, largely being responsible for causing (or at least being the transmission mechanism for) recessions. The housing bubble in the middle of the decade cased an enormous number of houses to be built, and a huge percentage of them were McMansions built in the ex-urbs, far from employment centers.
When the bubble popped, home prices plunged, and millions of people were left with houses that were worth less than what they owed on their mortgages. For a large number of those people, the most economically rational thing for them to do now is simply stop paying their mortgage and wait for the sheriff to knock on the door and kick them out. In most areas of the country, people can now live rent and mortgage free that way for over a year. However, eventually they will be kicked out, the banks will take over and want to sell the house.
This is the “shadow inventory” of housing, and it is huge — most likely totaling over 5 million houses. With so many excess houses on the market, it is not really economically rational to be building more of them, from a big picture point of view.
However, residential investment is normally what pulls us out of slumps. Each new house built creates an enormous amount of economic activity (used home sales, even though they are far more numerous, do not have anything close to the economic impact of new homes). Building a house employs lots of carpenters, plumbers and electricians. Those are relatively good-paying jobs for people who do not have a lot of formal education. They are also jobs that are very tough to outsource to China or India (although given the number of undocumented workers in the industry, some might say we “outsource” much of the work to Mexico).
Beyond that, building a house will generate a demand for wallboard from USG (USG), lumber from Weyerhaeuser (WY) and plumbing fixtures from Fortune Brands (FO). Those primary jobs then support other jobs as those workers then have enough money to go shop at Wal-Mart (WMT) or go out to eat at Pizza Hut, which is part of Yum! Brands (YUM).
Total housing starts, and especially single family starts are below the worst levels seen in previous recessions. The only other slump that comes close was the 1982-83 recession, and in that case the Fed was trying to wring inflation out of the system, and to do so they jacked up mortgage interest rates to almost 20%.
Now mortgage rates are at historic lows. The population of the country has also been growing at about 1% per year since the 1960’s and 1970’s. More people should mean more demand for places to live. The problem is that the rate of household formation is extremely low.
Household Formation
“Household formation” is economist speak for getting junior out of Mom’s basement and into a place of his own. It is also about getting people who have lost their homes and are now doubling up with friends or families into their own places. In both cases, the key to doing so is for those people to have a steady job so they can either pay rent or pay a mortgage.
But coming out of a recession, housing-related industries are normally one of the biggest engines of job creation. The construction industry was one of the hardest hit during the recession. It has lost 1.944 million jobs since the start of the recession, or more than one in four of all jobs lost in the downturn, even though at the start of the recession it accounted for just 5.46% of all jobs in the country. If those jobs are not coming back, then it is going to be tough to get household formation going, and without household formation, we will not absorb the existing oversupply of houses. Chicken meet egg; egg, chicken.
Building Permits
The best predictor of future housing starts are building permits (which incidentally telegraphed that we would have a downturn in starts this month by falling 11.5% in April, based on the original report). The permits data does not point to an early turnaround. This hangover is going to take more than a few cups of coffee to cure.
Total permits fell by 5.9% on the month to an annual rate of 574,000, and up just 4.4% from a year ago. They were also far below the consensus expectations of a 631,000 annual rate. In a small silver lining, the April data was revised up slightly to an annual rate of 610,000 from 606,000.
As with the starts data, strength in the very volatile multi-family area cushioned the blow. Single-family permits were down 9.9% on the month and are just 3.1% above the year-ago level. Permits for multi-family housing were up 9.3% on the month and are up 13.6% from a year ago.
Unlike with starts, all four regions saw declines on the month. Worst hit were the Midwest (-9.6%) and the West (-6.8%) the two regions that saw increases in starts in May. Thus it looks like they will catch up with the declines in the Northeast and the South next month. Permits in the Northeast fell 1.5% on the month and in the South permits were down 5.2%.
On a year over year basis, the Midwest is the strongest with a 7.3% increase, followed by the Northeast with a 6.3% increase and the South with a 5.0% year-over-year gain. Out West, permits are actually 0.9% lower than they were a year ago.
Building permits are an important leading indicator for the economy, and this continued weakness is not good news for the immediate future. Over the long run, not putting additional resources into building more houses when we already have so many of them probably makes sense. The lack of building activity (and it is not like commercial construction is picking up the slack) means that economic activity is going to continue to be extremely sluggish.
Job growth will not be strong enough to put a meaningful dent in the massive army of the unemployed. Without job growth, housing formation will continue to be sluggish, which means that the absorption of the excess housing supply will be slow. Eventually, population growth will cause the supply to be absorbed, and while we are not likely to see a return to the bubble level of activity anytime soon, it would not take very heroic levels to generate some very strong percentage gains. Just getting single family starts back to levels that would have been considered weak in the 1970’s or the 1980’s would result in more than a doubling from current levels.
When that starts to happen, it will be a sign that the private sector is picking up the torch from the government stimulus programs and we have an ongoing, self-sustaining recovery and expansion. Unfortunately, that does not look like it is going to happen in the immediate future.
This was a very weak showing in a very significant economic report, and points to slower economic growth in the second half of the year than in the first half.
Dirk van Dijk, CFA is the Chief Equity Strategist for Zacks.com. With more than 25 years investment experience he has become a popular commentator appearing in the Wall Street Journal and on CNBC. Dirk is also the Editor in charge of the market beating Zacks Strategic Investor service.
