It’s Housing Day over at Business Insider today, which is running duelling slideshows from Bill Ackman and Gary Shilling in the wake of yesterday’s dreadful home-price numbers. Both of them are a little dated—Ackman’s is from November 3, while Shilling’s seems to be from October, although it’s not entirely clear. Since then, prices have fallen, which would do little to change either of the analyses, but also interest rates have risen significantly, which puts a substantial dent in Ackman’s math.
Ackman’s logic is pretty simple: you can get a lot of leverage in the housing market, and if you make a highly-leveraged bet and prices rise, then you can get enormous returns. At the heart of his thesis is the idea that valuations are low and will rise in future—but to my eyes at least, there are problems with both legs of that argument.
The “valuation” section of the slideshow is a relatively small part of the whole. It says that home prices are 28% off their peak; that houses are more affordable now than they have been in a long time; and that buy-vs-rent calculations work out in favor of buying with relatively low home-price appreciation rates between 4% and 6%.
The first one seems meaningless to me: why should home prices be low at 28% off the peak, as opposed to say 18% off or 38% off? In fact, a glance at the Case Shiller graph shows that prices are still roughly double what they were for most of the 1990s.* And more generally, when the first part of a valuation argument is percentage-off-highs, I always get suspicious, because that’s always a really stupid metric on which to base a buying decision.
The affordability calculations, too, are a temporary phenomenon related to artificially low mortgage rates and the fact that the US government is currently providing substantially all housing finance. What I’m not seeing is any analysis by Ackman showing that houses will remain affordable in a future of higher rates and private-sector financing—the future when, I assume, he’s looking to exit his residential-property position.
Finally, required appreciation rates of 4% to 6% seem high to me, not low: they say to me that buying is still more expensive than renting unless you assume capital gains on your purchase. Ackman, here, comes close to assuming the very thing he’s trying to demonstrate.
As for the reasons for future gains in house prices, there’s a bit of basic demographics, in that the number of households in the US is going to rise over time, and is going to outpace the amount of new homebuilding. That’s reasonable.
Ackman’s other main argument is based on the assumption that homeownership is going to bottom out at 66% and stay there, rather than continuing on the long cyclical decline which started in 2004. If he’s wrong by just a single percentage point, a lot of his mathematics starts looking very shaky. I suspect that as America moves out of the suburbs and exurbs and into denser forms of living, we’ll see a steady decline in homeownership, which in any case is now seen as being more of a liability than an asset. It’s no longer a road to riches: it’s a road to possible foreclosure, bankruptcy, and an inability to move to where the employment opportunities might be.
Ackman is a speculator, and he’s forecasting a return of his speculative mindset among the population at large—where people buy homes because they’re confident in the future and think that those homes will rise in value, and also because they fear not being able to afford a house in future. That mindset fuels housing bubbles, not sustainable home-price appreciation. And investing in bubbles is always very dangerous, and generally ends in tears. Maybe Ackman can make money doing it—but that certainly doesn’t mean that homes are a good buy right now for the rest of us.
Ackman does have one intriguing idea about what might drive house-price appreciation: institutional investment in single-family home rental properties. He’s right that such things barely exist as a financial asset class, and that even a small global allocation to them could change the demand dynamics substantially. But being a landlord is hard work, especially in suburban and exurban neighborhoods, and I’m not convinced that it scales well: my guess is that the cost of hiring a good rental agent is always going to wipe out returns, and that the only way to make money by renting out single-family homes is for the owner and the rental agent to be the same entity.
If you then turn to Shilling’s presentation, things start looking positively depressing: he brings up reasons for pessimism that many of us never even consider, along with the much more obvious ones. At the top of the list, of course, is unemployment, which poisons everything. No one wants to buy a house if they don’t feel secure in their job, and people don’t feel secure in their jobs if unemployment is over 9%, where it’s likely to stay for the foreseeable future. And these graphs just speak for themselves:
This last one is the most powerful, to me: for all the wheel-spinning that we’ve seen of late in the mortgage-refinance market, the number of people shopping for new homes has been declining steadily for the past five years. I’m sure that Bill Ackman would look at that chart and extrapolate all manner of fabulous mean-reversion, but I can’t see where the new demand is going to come from. Buying a home is no longer the American Dream; people are much more interested in being free of debt. And while Ackman might look at the enormous leverage in the housing market and see opportunity, most Americans I think are now much more aware of the downside risk.
No matter what happens in the housing market, Bill Ackman is always going to have more wealth than he can ever spend in his lifetime. But for those of us without his ability to make large speculative bets, housing is a very scary asset class, while renting is cheaper and much more flexible. I wouldn’t be at all surprised to see the US homeownership rate fall a lot in coming years, back down below even its long-term mean around 64%. And if that happens, prices—both to rent and to buy—are almost certain to fall from current levels.
*Update: Thanks to ckbryant, in the comments, who points out that this is true only in nominal terms, not in real terms.