“Hope meets Reality – A Slow March to Recovery”
By Mike Tonti
U.S. housing starts reported a higher-than-expected increase in January which is a positive sign for the housing market to begin the year. Total housing starts jumped 14.6% from December levels to a seasonally-adjusted annual rate of 596,000 units in January.
Monthly housing starts is a typical media driven headline with little to no meaning as the coincidence that a bunch of buildings started during a particular month or period of time reflects weather, labor, commodities, debt pricing, when you need to deliver the product and the start of large projects. Building permits represent the real trend or pipeline of units which we will see in the near future (1 to 2 years) as the permit process takes anywhere from 120 days (for single family) to three years (west and east coast multifamily). Annual housing starts peaked in January 2005 at 2.068 million starts. We are thus still 73.8% off the peak levels. Single-family starts are 77.3% below peak levels. Construction jobs represent 25% of this recession’s job losses forcing many housing workers to leave the industry until the U.S. absorbs the excess housing units. Eventually equilibrium will be met, and the demand will again exceed supply. At that point prices will rise for existing housing as well as workers as the industry attempts to bring workers back into housing production to meet demand.
Look for lots of headlines about housing as an indicator that the economy is recovering. In reality, housing will most likely be a lagging indicator due to record overbuilding during the boom. In the meantime, we will continue to look for properties positioned to benefit from a rebound. This will first occur within particular submarkets (cities such as Austin) until eventually enough submarkets recover to see a stable national housing market and a recovery underway.
How long will this take? No one really knows but we peaked in 2005(in home ownership and starts) with the dramatic drop in deliveries in 2008, 2009, and 2010, see the chart below.
Annual Housing Starts (1978-2010)
| Year | Single-Family | Multifamily | Total |
| 2010 | 470,900 | 116,700 | 587,600 |
| 2009 | 445,100 | 108,900 | 554,000 |
| 2008 | 622,000 | 283,500 | 905,500 |
| 2007 | 1,046,000 | 309,000 | 1,355,000 |
| 2006 | 1,465,400 | 335,500 | 1,800,900 |
| 2005 | 1,715,800 | 352,500 | 2,068,300 |
| 2004 | 1,610,500 | 345,300 | 1,955,800 |
| 2003 | 1,499,000 | 348,700 | 1,847,700 |
| 2002 | 1,358,600 | 346,400 | 1,704,900 |
| 2001 | 1,273,300 | 329,400 | 1,602,700 |
| 2000 | 1,230,900 | 337,800 | 1,568,700 |
| 1999 | 1,302,400 | 338,500 | 1,640,900 |
| 1998 | 1,271,400 | 345,500 | 1,616,900 |
| 1997 | 1,133,700 | 340,300 | 1,474,000 |
| 1996 | 1,160,900 | 315,900 | 1,476,800 |
| 1995 | 1,076,200 | 277,900 | 1,354,100 |
| 1994 | 1,198,400 | 258,600 | 1,457,000 |
| 1993 | 1,125,700 | 162,000 | 1,287,600 |
| 1992 | 1,029,900 | 169,900 | 1,199,700 |
| 1991 | 840,400 | 173,500 | 1,013,900 |
| 1990 | 894,800 | 298,000 | 1,192,700 |
| 1989 | 1,003,300 | 372,900 | 1,376,100 |
| 1988 | 1,081,300 | 406,700 | 1,488,100 |
| 1987 | 1,146,400 | 473,800 | 1,620,500 |
| 1986 | 1,179,400 | 626,000 | 1,805,400 |
| 1985 | 1,072,400 | 669,500 | 1,741,800 |
| 1984 | 1,084,200 | 665,300 | 1,749,500 |
| 1983 | 1,067,600 | 635,500 | 1,703,000 |
| 1982 | 662,600 | 399,700 | 1,062,200 |
| 1981 | 705,400 | 378,900 | 1,084,200 |
| 1980 | 852,200 | 440,000 | 1,292,200 |
| 1979 | 1,194,100 | 551,100 | 1,745,100 |
| 1978 | 1,433,300 | 587,100 | 2,020,300 |
We do not seek to pick the bottom, but to invest along the curve if the opportunity hits our investment criteria. The dollar cost averaging approach to investing in housing following a cyclical overbuilding will allow the investor to average in on a cost basis and capitalization return. For example a rising capitalization rate market over three years from 7% to 8% to 9% would leave an investor with a 8% average return and basis. The investor benefits from earlier acquisitions which have locked in lower debt rates while still participating in higher capitalization purchases/debt later. The opposite is true if we face deflation, so laddering is tactically the best method to address interest rate risk. The value investing timing is based on the overbuilt housing industry recovering over time at a rate relative to both its submarket and the overall national economy. Nonsystematic risk events will interrupt the rate of recover/timing but ultimately equilibrium will be reached. We continue to believe this is a good time to invest in housing.
