I was recently reading an article that mentioned prospects in the housing sector and suggested D.R. Horton, Inc. (DHI) and Lennar Corporation (LEN) as possible longs. I've not looked at the residential construction market for a while, but it struck me that the time to buy these companies has largely passed. Furthermore, it seems that relying upon the government issued data probably results in too much lag. The U.S. Census Bureau's most recent residential construction related data is from November 2012. Stock prices are reflective of investors' expectations for future performance and common valuation approaches are based upon discounted future cash flows or some variation on that.
Housing starts are still at historical lows
Looking at data shows that the housing market is still way below the historical average for starts. The following graph shows the annual housing starts going back to 1959.
Source: U.S. Census, 2012 data is based upon 11 months giving 720 thousand starts, my personal estimate would be 773 thousand starts for the full year.
The long run average was around 1.5 million per year. However, this is slightly misleading since one would expect housing starts to be related to population or existing housing stock. Over time, one would expect a relative upward trend assuming that the average life of a home remains constant. That upward trend is not immediately discernible in this case. However, in any case, the five years in recorded data with the fewest starts are 2008-2012, with 2009 as the lowest with only 554,000 starts. The November 2012 (annualized and seasonally adjusted) result was a paltry 861,000 starts. 2005 saw the most housing starts (~2.1 million) since the peak in 1972 (~2.3 million).
Housing stocks have rebounded
In 2005, housing starts were at their peak. One would reasonably conclude that expectations were also going to peak somewhere around that time, but most probably lagged. I would wager somewhere in 2006. On an adjusted basis, January 2006 was the peak month, with that data becoming available perhaps in March 2006 and then at perhaps several more months before any one could wager that there was a clear downward trend occurring.
|Type||Ticker||Name||Recent Price*||NA/MC ($Millions)|
|Stock||DHI||D.R. Horton, Inc.||19.78||6,353|
|Stock||TOL||Toll Brothers Inc.||32.33||5,438|
|ETF||XHB||SPDR S&P Homebuilders ETF||26.60||2,240|
|ETF||ITB||iShares Dow Jones US Home Construction ETF||21.16||1,610|
Source: Yahoo Finance, Stock price as of close on December 31, 2012. NA is net assets for ETFs or MC (market capitalization) for stocks.
The following graph will compare the monthly stock prices of some of the aforementioned ETFs and homebuilders.
Source: Yahoo Finance, U.S. Census, Author calculations. Stock price data is adjusted for dividends.
This graph shows a few things amongst the colored squiggles. I will mark select dates for each and show the normalized data for all 6 stocks and ETFs in tabular form:
- TOL "peak" - June 2005
- Housing stocks peaked before the broader market as measured by SPDR S&P 500 Trust ETF (SPY) and was in decline by the time SPY peaked. Housing stock "peak" - January 2006
- SPY peaks - October 2007
- Housing starts bottom at just under .5 million seasonally adjusted annualized starts - April 2009
- Fall of 2011 anxiety over - September 2011
Source: Yahoo Finance, U.S. Census, Author calculations, ITB and XHB were not established until May and Feb of 2006 respectively. Starts SA (Seasonally adjusted) for Dec-12 is actually the November 2012 result.
This data suggests that there has already been a pretty strong recovery for many housing related securities coming off their lows as expectations and market conditions have continued to improve. It should also prove cautionary to those hoping for future outsized returns. Supporting the first point - an investor with the stomach to purchase LEN during the depths of panic in April 2009 has now had an approximate 300% return. If LEN returns to its price in the housing euphoria, one would only get another approximate 50% return. While 50% is a nice return, the question is will we see the same type of 2006 housing euphoria any time soon? Similarly for ITB, which is at just 52% of its Jan 2007 level and shows approximately 93% return from April 2009. Investors could get maybe another 100% it goes back to the January 2007 level. (100/52 -1 = 92%). So again in this more extreme case, already half the potential return has been captured. TOL is essentially where it was six years ago.
At this point, housing stocks require caution
The recovery in housing stocks has been much stronger than the recovery in actual housing starts, emphasizing the point that stock prices are based upon future expectations. This compares the 61% figure to all the other December 2012 figures. In general, the stocks have recovered more. Furthermore, many housing stock investors have already made some outsized returns. For every share of stock or ETF that was sold in April 2009, there was a corresponding buyer who is now quite happy. Furthermore, one can see the stumble in the housing stocks in September 2011 despite the relatively stronger broader market. Based on price levels, September 2011 was better for the broader market than April 2009, while the reverse was true for most of the housing stocks and ETFs.
Looking forward in the near future (1-2 years), I don't see outsized returns for housing stocks in general and hence I don't see it as a great opportunity. However, this is not to say that there will not be winners and losers within the sector. One could still potentially make a good return on picking within the sector. This article makes little distinction between companies and does not seek to identify strengths and weaknesses (valuation, management, strategy, geographic focus, etc..) of individual companies. However, there is an intriguing notion, that some individual companies could do very well. Looking at housing starts relative to the long term mean suggests there is significant future opportunity that some companies will capture and thus realize the expectations in their price. The first graph that I showed is still quite startling and raises the question of whether we will revert back to some mean or is there a new paradigm in play and for how long.
My other observation is that the two ETFs analyzed were formed just in time for long investors to get clobbered. Perhaps this is a bearish indicator for future reference.
Disclosure: I am long SPY.
Additional disclosure: Disclaimer: This article is for informational and educational purposes only and shall not be construed to constitute investment advice. Nothing contained herein shall constitute a solicitation, recommendation or endorsement to buy or sell any security.