Why Homebuilders are in Trouble
Since the financial crisis in 2008, the housing sector has been leading the US economy back out of the recession. As shown below, the XHB (SPDR Homebuilder's ETF) has more than tripled since 2008. The homebuilding sector has benefited from the credit easing policies of the Federal Reserve, as well as from a large increase in home prices. However, with the Federal Reserve's announcement in May that it is considering reducing bond purchases and with the size of its balance sheet now over $3.7 trillion dollars, consensus is that the Fed will begin to taper back its bond purchases in the upcoming months. Also, existing home inventories are up 28% YTD. This should negatively affect home prices, and should begin to show itself in the lagging housing data. Hence, I now believe the risk-reward ratio is compelling to short the Homebuilding sector.
Some (Recent) History
As shown below the Homebuilding sector (XHB) has increased by over 275% since the depths of the 2008 financial crisis. A large part of this movement has been due to the Federal Reserve's dovish policies. Not only did the Fed reduce the Fed Funds Rate to 0% in order to inspire more banks to lend, the Fed began to directly purchase both Treasuries and Mortgage Backed Securities. This artificial demand has led yields on bonds to fall, forcing banks to increase lending and allowing consumers to finance (or refinance) their mortgages at lower rates. Another reason the homebuilding sector has performed so well has been due to an increase in home prices. This has been led by existing home inventories falling by over 34% since 2008.
The Tide Has Begun to Turn on Rates
The Fed announced in May that it might begin to scale down the size of its purchases of Treasuries and MBS over the next few months. The markets reacted. 10-year Treasury rates went from under 2% to 3% and a 30-year fixed mortgage rate increased from 3.3% to over 4.3%. The intuitive inverse correlation between mortgage rates and mortgage purchases applications is shown below. According to ISI, the increase in mortgage rates has caused monthly payments to increase by over 35%. Most important, the Fed decided not to taper in part due to a bad housing number that came in between May and September. From Chairman Bernanke, "the tightening of financial conditions observed in recent months, if sustained, could slow the pace of improvement in the economy and the labor market." However, when the Fed announced that it was not going to taper 10-year Treasury rates only fell from 3% to 2.7%, meaning credit is still tighter than it has been for the past two years and the fact yields did not fall back to May levels shows that the Fed is losing control of the market.
Rates have never gone up so fast (And stayed up!)
Never in history have yields on 10-year Treasuries sustained a 35% increase (2.7% from 2%) for multiple months. Therefore, it is hard to judge how large of an impact higher rates will have on the demand for housing. That being said, we can point back to when the $8,000 Obama home buying tax credit expired in April of 2010. As shown in the image below, the effect of the tax cut expiration was gigantic.
The average of Mortgage Purchase Applications 6 months leading into the tax credit expiration was about 230/week. After April, the average dropped to around 185 /week. While there is no precedent for sustained higher rates, the expiration of the tax credit points to the conclusion that sustained higher rates may have a large negative impact on demand.
So, there is a correlation between Mortgage Purchase Applications and Mortgage Rates. But not Mortgage Purchase Applications and New Home Starts…
You would assume that Mortgage Purchase Applications and New Home Starts would be highly correlated - and they have been in the past. However, what is confusing is the disconnect between the two starting in 2011. My thoughts are that the disconnection of the correlation is due to either 1) People downsizing and paying for a new home with cash. 2) Investors are entering the market as a result of REITs or simply people are deciding to buy a house and rent it out. Either way, I look at this as the housing sector taking on more leverage. Leverage not in the sense of borrowing, but in the sense that the fundamentals of the housing market are changing because investors have been pushed into investing in the housing market. Thus, when the underlying fundamentals begin to get worse (lower prices and less 'real' [MBS] demand) investors will try to leave the market. This could result in the housing 'bubble' bursting.
Housing Prices will Begin to Fall
My theory is that home prices have been increasing since 2008 due to lower supply, not higher demand. As shown below, there is an inverse correlation between the Case-Shiller Home Price index and existing inventories sales. However, since the beginning of 2013, inventories have increased by 29%. This increase in inventories should begin to show itself in the Case-Shiller index in the next few months.
The Case-Shiller Index
The Case-Shiller index is the most trusted index for finding home prices in the United States. Unfortunately, the Case-Shiller index is a 3-month moving average of home prices and it records its pricing data from closings. Closings are often 8 or more weeks after a bank and a consumer agree on a mortgage. On top of that, consumers often file for Mortgage Purchase Applications several weeks before. Thus, in total, mortgage purchase applications lag the index by roughly 5 months. (For more see here.) If you match the durations of the Case-Shiller index with the Mortgage Purchase Application, you get an inverse correlation of -.75. Thus, my view is that housing prices in the Case-Shiller index will begin to fall as a result of more supply in the market.
Lumber Price Disconnection?
As shown below, there has been a high correlation (.75) between lumber prices and housing stocks. However, recently there has been a huge disconnect as lumber prices have tumbled, while housing stocks have remained at the same value. Lumber prices have recently rallied some, which may make it seem that maybe lumber was 'wrong' and housing stocks were 'right', but nonetheless the disconnect is concerning.
How do we play this?
I had trouble finding data on the XHB etf. However, according to Yahoo! Finance, the XHB has a ttm P/E of 18 versus a 16.96 P/E on the S&P 500. Moreover, with fundamentals deteriorating, I recommend shorting the XHB (either outright or against the S&P 500 depending on your market view). Below shows the effects that multiple compression would have on the price of XHB. The bolded values are the values that the XHB would be trading at if it had the same multiple as the S&P 500 or had a multiple of 10% lower.