By Jamie Howsley
Fall 2012
A few years ago I wrote an article entitled "Demographics, Development and Demand," in which I predicted a return of the housing market by late 2010. Well we all make mistakes. But the underlying theory behind the article is now taking shape. Namely, that because over the past several years housing starts fell well below the historical average of approximately 1,544,000 units a year built between 1978 and the end of 2007, a return to near normal production had to occur at some point simply to satisfy the need of new household formations.
According to the U.S. Census Bureau, nationwide we saw approximately 906,000 units in 2008, 554,000 units in 2009, 587,000 in 2010, and 609,000 units in 2011. According to data released by the Census Bureau on September 19, 2012, housing starts are at an adjusted annual rate of approximately 750,000. This is a five-year average of approximately 681,000 units a year, well below the 1,544,000 unit average referenced above. If we accounted for the overproduction that may have occurred during the robust cycle of 2003-2007, the underproduction of 2008 and 2009 should have normalized the numbers' historical average already.
But a weak economy continued to persist during the remainder of 2010 and all of 2011. While many economists declared the end of the "great recession" some time ago, it continues to linger.
In the past two to three months I have noticed an emerging trend. Builders and developers dormant for some time have begun calling again not just to chat, but to inquire about land that had only a preliminary plat on it or even raw land. Why is this important? Conventional wisdom would suggest a severely dwindling lot inventory, the very thing that I warned about in that 2010 article.
The warning to the real estate community may seem a bit strange given the state of the market in the past few years. But if housing production doesn't return to the historic norms mentioned above as soon as the economy has fully recovered, we will be facing a severe housing shortage that could cause housing and lot prices to reach the fevered pitch they had during the height of the real estate boom. Because land is a commodity in the development world, builders and developers would be out there once again competing for an ever-decreasing supply of lots and land.
This follows some words of advice a developer client of mine gave me years ago. He had seen several cycles and said that things would remain sluggish until 2014 and then, look out! He knew, and still knows, how long it takes to bring new lots to market. In the Portland/Vancouver metropolitan market it takes at least two years, and in some cases four or more, to bring raw land through the approval process and have the infrastructure in place to begin constructing and selling new platted lots.
According to New Home Trends' September 2012 data, since 2007 there have been about 7,800 new lots brought to market while there have been about 9,700 new homes sales during that same period. This would suggest that the supply of lots should be extremely low, even if a glut of additional lots were created prior to 2007. And the rate at which new lots have been brought to market over the past few years is also telling. In 2009, it totaled around 839 lots, in 2010 642 lots, and in 2011 526 lots.
During the downturn, the negative news in real estate artificially suppressed developers from seeking approval of new subdivisions and apartments. Even when a developer wanted to pursue new approvals it would be stymied by strict lending requirements, all of which created an odd paradox of sorts. Banks were told by the Federal Deposit Insurance Corporation that they needed to reduce their real estate holdings and hence began selling lots. The FDIC forbade many banks from lending on new projects despite there being an eventual demand for them in the coming years. Yet now we are on the edge of a shortage.
I am hopeful that we will continue the upward trends, but we should remind ourselves that we could slide back to some very frantic times that could once again produce negative results.
