By James Kwak
Loyal readers already know what I think of housing as an investment. The main issue, in my mind, is that it’s extremely risky as an investment: not only are most middle-class families putting more than their total net worth in a single asset class (and one with low average real returns compared to the stock market), but they are putting it into a single asset, which violates the most fundamental principle of investing.
That said, on a pure expectation basis (not considering risk), buying is probably better than renting. It’s not as simple as saying that “renting is throwing money away while paying a mortgage is building equity” because (a) homeowners usually pay more cash than renters on an ongoing basis (mortgage, homeowner’s insurance, maintenance, etc.) and (b) you have to consider the returns you could get by investing your capital (down payment and principal payments) in another asset class. But the tax deduction for mortgage interest probably tilts the scale toward buying.
So if you’re thinking about buying or renting, I recommend that you read “The Effectiveness of Homeownership in Building Household Wealth” by Jordan Rappaport, an economist at the Kansas City Fed (hat tip David Leonhardt). The most valuable part of the paper is that it clearly outlines the financial tradeoffs between owning and renting. Rappaport creates a model that estimates the cash flows from buying a house and selling it ten years later and renting for ten years, assuming that you invest all the money you save by renting. He then looks at historical ten-year periods beginning from the 1970s through the 1990s to see which strategy would have been preferable.
It turns out that, of those ten-year periods, owning was better in about half, renting was better in about a quarter, and in the other quarter it’s hard to tell. The ten-year periods that began around 2001 will probably favor owning, but when you get to 2003-2006 they will probably favor renting (because of the housing decline since 2006).
So, in short, owning does better than renting somewhat more often than renting does better than owning. Does that mean that buying a house is a good investment? It depends. First of all, Rappaport’s model assumes a ten-year holding period; as your expected holding period decreases, owning becomes less attractive because you have less time over which to amortize the transaction costs.
More importantly, Rappaport’s conclusions are based on averages — in particular, average house price appreciation. This goes back to my original point: housing is a lot riskier. Your house’s price appreciation could deviate wildly from the average in your neighborhood, let alone your Metropolitan Statistical Area, let alone the country, for any number of reasons. So the question is how much investment risk you want to take on. Sure, rental prices could go up faster than the national average. But over the ten years, an owner will be more sensitive to a fall in housing prices than a renter is to a rise in rental prices of the same annual percentage. (And that leaves aside the fact that a renter can adapt to unfriendly price changes — by moving — more easily than an owner can.) For a good discussion of the specific risks of owning, see Rappaport’s discussion on pages 45-46.
So if you are making the decision for real, the tradeoff is this: historically, on average, owning has beaten renting more often than the converse (if you have a ten-year holding period), but owning is riskier. You can’t tell now what is going to happen to house prices over the next ten years. But there is one thing you can sort of estimate: the current ratio of house prices to annual renting costs. David Leonhardt‘s rule of thumb is that if the ratio is below fifteen, owning will probably work out better than renting, and if the ratio is above twenty, renting will probably work out better than owning. (That is roughly borne out by Rappaport’s chart on page 51, but I think the chart implies that the “owning” threshold should be somewhat lower than fifteen.)
Now for the usual caveat: Everything above discusses buying a house as an investment. Most people get more consumption value from owning a house than from renting an equivalent house because most people get utility from living in a place that they own. They like the security, the ability to make modifications to the house (even though most of those modifications are probably money-losers), and so on. Also, in many markets you just don’t have a choice. I live in a small college town, and the rental market here is primarily geared toward students, so it’s hard to find a nice house for rent (especially one that will let you have a dog, which we did when we moved here). So there are plenty of good reasons to buy a house other than the idea that it’s a good investment. Those are valid reasons why you might buy a house even while thinking that it’s a bad investment. Except for scale, it’s no different from, say, eating at a nice restaurant now and then. Buying an expensive dinner is a lousy investment, but it gives me consumption value.
As Rappaport says,
“Homeownership, at least until recently, was often described as a great investment that also includes a place to live. More accurately, homeownership should have been described as a place to live that also includes an expected investment benefit.”
There is an expected benefit there, but remember that it’s not that big and it comes with greater financial risk.