Housing Starts fell in April to a seasonally adjusted annual rate of 523,000 from 585,000 in March, a decrease of 10.6%. However, the March numbers were revised sharply higher, from 549,000, so it is possible to see the decrease as 26,000, or 4.7%. Relative to a year ago, they are down 23.9%.
It’s not that things were going great guns for the homebuilders last year, but sales were goosed by the end of the homebuyer tax credit. Thus the year-over-year decline is off of a relatively tough comp.
If one looks at only single-family houses, the picture was not quite as bad, but still not pretty. Single-family starts fell to 394,000 from 415,000 in March, a fall of 5.1%, and down 30.4% from a year ago. March starts were revised down from 422,000, so single family starts are down 6.6% from where we thought they were last month.
The volatile multi-family (Apartment, Condo and Co-op) sector plunged by 28.3% to an annual rate of 114,000 from 159,000. Year over year, multi-family starts are up 5.6%. The first time buyer tax credit was really not a factor in the multi-family starts a year ago. The total starts number was well below consensus expectations of a 565,000 annual rate.
In any absolute sense, the level of housing starts is just plain awful. The extremely weak rate of new home construction is a major drag on the economy. It is the principal reason that this recovery feels so anemic. The silver cloud is that fewer starts means that there are fewer houses added to the inventory of houses looking for buyers.
We still face an inventory glut, so a weak homebuilding industry is a key part of the repair process for the housing market. The inventory glut is concentrated in the used home segment of the market, and that is also where the “shadow inventory” resides. New home inventories are actually near historic lows in absolute terms. Used homes are pretty good substitutes for new homes, so that is a bit of a distinction without a difference.
Housing Starts peaked in June of 2006 at an annual rate of 2.273 million. We are thus 77.0% off of the peak levels. Single-family starts are 78.5% below peak levels.
The decline this month will speed the adjustment process, but also means more economic weakness over the next few months. On balance, I have to see the increase in starts as being a bad thing. We need the economy to get moving again, and add more jobs. With more jobs will come a higher demand for housing. However, we don’t want to see housing starts rise without new home sales also rising, otherwise we will just be exacerbating the overall inventory problem.
Housing Hugely Important
It is hard to overstate just how important housing starts are to the economy. Yes, at this point, residential investment has declined to the point where it looks almost insignificant, just 2.21% of GDP in the fourth quarter, down from 6.34% of GDP at the height of the housing bubble. However, historically, residential investment, of which new home construction is the largest part, has always been the main locomotive in pulling the economy out of recessions.
Take a good hard look at the first graph below (from http://www.calculatedriskblog.com) and the relationship between when the lines bottom and the light blue recession bars. If you want to know why this recovery seems so anemic, look no further than this graph. Even the 2001 recession, which was not caused by a housing downturn, saw a sharp acceleration in housing starts as the recession came to an end.
Of course, since starts were jumping but were not starting from a depressed level, that boom later became known as the housing bubble that put us in this mess to begin with. Every other recession was preceded by a sharp fall in housing starts, every other recovery saw housing starts lead the way.
This is no coincidence. Each new home built generates a huge amount of economic activity. It put construction workers back to work, and construction workers have been particularly hard hit in the Great Recession, accounting for over 25% of the total jobs lost, even thought they were less than 6% of the total workforce when the recession started. That is just the direct construction jobs, but lower building activity also means fewer jobs in the factories that produce building materials, which are counted as manufacturing jobs.
Clearly jobs in mortgage finance are also affected by the housing slowdown. They are not included in that “one out of four jobs lost” figure. As they and the construction workers go back to work they are also going to have more money to spend, perhaps even go out to eat, thus creating jobs for cooks, waitresses and busboys.
Housing starts are not just about profits and jobs at D.R. Horton (DHI) and the other homebuilders, but about jobs and profits at firms as diverse as Plum Creek Timber (PCL), Fortune Brands (FO) and Berkshire Hathaway (BRK.B). Indirectly, it even helps Wal-Mart (WMT).
Why is housing so central, as opposed to other industries? Why does it hold the key to the economy booming or busting? Because it is exquisitely sensitive to interest rates, or at least it was, before the avalanche of houses in foreclosure simply swamped the housing market.
Even near-record-low mortgage interest rates don’t seem to be moving the needle. The simple fact is that during the housing bubble we built far too many homes, and we now have a glut of empty homes around the country. Most estimates put the excess vacancies at between 1.2 to 1.5 million (including rental units).
The next graph (also from http://www.calculatedriskblog.com) shows the total (homeowner and rental) vacancy rate over time, relative to housing starts. While it is off its peak, it is still far above normal. In such a situation, it seems economic folly to simply build more houses and add to the glut. But if we don’t build houses, the economy remains stuck in a rut. From a strict “allocation of resources” point of view, we would want to see slow housing starts until the vacancy rate fell back to more normal levels (say, under 4.0%).
Thus, one can argue that in the long term, low housing starts are a good thing, as it means fewer new homes adding to the glut. That, however, is extremely cold comfort to the millions of construction workers who are out of work. It is also going to be very difficult to create a sustained growing economy if home building continues to be a drag.
Residential investment has been a relative non-factor in the last two quarters as far as overall economic growth is concerned, adding 0.07 points to growth in the fourth quarter, then subtracting 0.09 points in the first quarter. However, over the past few years it has been a persistently negative force on the economy.
The upward revision for March suggests that Residential Investment will be a more positive contributor to first quarter GDP growth in the next revision, although not a huge one. The weak starts in April suggest that the second quarter is not going to get much help from Residential Investment, and as a result, could well be another sub-par quarter for growth.
Results by Region
For the month, the results were very mixed by region. Two of the four census regions posted increases. The Midwest was the strongest, with total starts rising 15.7% on the month, but down 19.3% year over year. The West was also up, with a 3.7% increase, and was the only one to post an increase year over year, up 2.8%.
The Northeast posted a drop of 4.8% for the month, and is down 29.4% year over year. The South, which is by far the largest of the four regions, accounting for 48.8% of all starts for the month, was the weakest by far. For the month starts were down 23.0% and are off 31.8% year over year.
The decline in starts is discouraging, and unfortunately, it is likely to continue. The best leading indicator of housing starts are Building Permits. There the news was also bad, at least from a near-term economic growth point of view. (Or good from a long-term repair of the housing market point of view).
For the month, total permits fell 4.0% to an annual rate of 551,000 and were down 12.8% year over year. That was well below the consensus expectations for a 590,000 rate. Compounding the blow, March was revised down, from a 594,000 rate to 574,000. The weakness was mostly in the multi-family part of the market while the drop in single-family permits was more modest.
Single-family permits were down 1.8% on the month, and down 18.6% year over year. Multi-family permits plunged 13.9% on the month but are up 2.1% year over year.
Regionally, the Northeast was the strongest for the month, with permits unchanged on the month but off 16.7% year over year. The West saw permits fall 0.8% for the month and off 3.9% from a year ago. The South, the largest of the four regions, fell 5.7% for the month and down 11.7% from a year ago. The Midwest was down 5.3% on the month, and down 23.3% year over year.
A Fall in Housing Starts – Good or Bad?
There is a certain level of ambiguity about if a fall in starts is good news or bad. We need less supply to help the market clear, but in the meantime, the economy is going to be stuck in limbo, unless we can find another locomotive to help pull us forward. The one we have always relied on in the past is clearly derailed.
Every housing start represents a lot of economic activity, and the effects go far beyond the bottom line of the big home builders. Housing starts are an important part of the job picture, and construction has historically been an important source of relatively high paying jobs for those without a lot of formal education. The conundrum is how to get these people back to work without simply adding to the existing housing glut.
Additional government spending on infrastructure would be the logical solution. After all, we have huge needs to rebuild out crumbling infrastructure. Better infrastructure would enhance our long-term economic competitiveness. The government would not be competing for resources with the private sector — it would be competing with idleness.
The cost of financing the infrastructure rebuild would be extremely low, give that T-note yields are in any historic perspective, low across the yield curve. Of course, we can’t borrow to fix our crumbling bridges, water treatment plants and roads unless the debt ceiling is increased.
Housing Will Come Back (Eventually)
Eventually, the combination of a rising population and higher household formation will absorb the excess housing inventory, and we will be able to once again increase housing construction. “Household formation” is economist-speak for kids moving out of their parents houses and getting a place of their own. To do so takes a job, and one that pays enough to support having your own place.
In the past, residential construction itself provided a lot of those jobs. That produced an upward spiral. This time around, the jobs have to be created in other parts of the economy to increase the household formation rate, and absorb the excess inventory, which makes the process slower. This process should probably hit the rental market first, and we are starting to see that.
Most of the decline in the vacancy rate shown above has come from the rental side of the housing market. After all, kids moving out of Mom’s basement are more likely to first move into a rental apartment than to buy a house. The decline in the rental vacancy rate is thus a positive omen for the future.
The day will come when housing is once again a big positive contributor to economic growth. It is not really sustainable over the long term to have total housing starts running at substantially lower levels than we saw in the 1960’s. Back then, the population of the country was around 200 million, now it is north of 310 million.
That day, however, is not today nor in the near future. It is possible that it might happen in the second half of 2011, but more likely in 2012. For the time being, the best we can really hope for is that it stops being a brake on the economy. That appears to be happening.