Your study looked at the annual cost of owning a single-family home for 46 U.S. markets from 1980 to 2004. When you began the study, were you expecting to find evidence of a bubble?
Yes. I think, like most people, when I looked at the housing market it was puzzling to see prices rising so quickly, and I didn’t quite understand the process of why that was. But the more I looked at it, the more I realized that this was really what a fundamentals model would predict for housing.
The big insights of the study are severalfold. The first is that as interest rates get low, the impact of a change in interest rates gets bigger and bigger. The reason we’ve seen house prices grow so rapidly over the last six or seven years is that rates have been declining, and the lower the level, the bigger the percentage effect of a decline in interest rates on house prices.
The second insight — which is, I think, equally important — is that changes in interest rates have different effects depending on the market you’re looking at. In particular, supply-constrained markets, like in the Northeast and California, are markets where the price-to-rent ratio is already pretty high, and changes in interest rates have a bigger percentage effect on prices than in markets in the South and in the Midwest, where the growth rate of prices is just inherently slower. That means we should expect — in the same sense that the price of Google stock is more volatile than the price of a slower-growing company — that the price of housing in some of these fast-growing markets is going to be more volatile, both on the upside and on the downside.
The third insight is recognizing that the price levels in most of these markets are not excessively high. If you look at the cost of owning a home, which takes into account changes in interest rates as well as appreciation, maintenance and other factors like that, the cost of owning a home is only modestly higher than the historical average over the last 25 years. The cost of owning a home has gone up a lot in the last 10 years, but 1995 was probably the best year in the last 25 to have bought a house, so people who bought houses at that point have gotten a pretty big windfall. But that doesn’t mean it’s a terrible time to buy a house today. It just means we shouldn’t expect that kind of a windfall in the future. I don’t think people should buy a house expecting to see the kind of appreciation they’ve had, but it doesn’t mean that their house price is going to collapse, either.
Can you talk briefly about your research methodology?
We argue in the study that the right way to look at the price of housing is really to look at the after-tax cost of owning a home. That starts with where interest rates are, and our view is that it’s really important to look at long-term, not short-term, rates. So the big run-up in short-term interest rates in the last year and a half doesn’t matter as much as the fact that long-term real interest rates are very low. So we start with long-term mortgage rates and then look at the tax benefit associated with owning a home, and then parcel out changes in property taxes, in maintenance costs and features like that.
Finally, the last piece of the cost of owning a home, of course, is the expected appreciation or capital gain you get from a house. And here, to be conservative, we’ve not assumed that people are going to get the most recent run-up in house prices, but we’ve assumed that they would get the average after-inflation increase in house prices over the last 60 years. The idea is to look at a very long-run measure, not to build a bubble into the model, and then ask, given that very long-term appreciation, what is the cost of owning a home?
And so we calculated that cost out for each of 46 markets in the United States. When we look at those costs, I think we find that a few markets tend to be high priced, but not the markets that people are used to seeing. The three markets that look most pronounced in the study are South Florida, particularly Miami and Fort Lauderdale; Southern California, in particular San Diego; and Portland, Oregon. Markets like San Francisco and New York, which people have focused a lot of attention on, do not show up in our study as having prices that are unfounded by fundamentals. And I think that’s a big observation for us, relative to what one reads in the popular press on housing.
You concluded that even in a market like San Francisco, which has an unusually high price-to-rent ratio, owning is not necessarily more expensive than renting. Why?
San Francisco house prices have gone up about 2 percent faster than the national average for 60 years. That means that part of the reason you’re paying a high price relative to the rent is that somebody who owns a house in San Francisco over time is going to get a big capital gain on their house. And so they should pay more for the house, given the capital gain that they’re going to get.
By contrast, if you bought a house in Houston, that’s a market where after-inflation house prices have fallen since 1975. If you buy a house in Houston, you should not buy it expecting it to go up in value. Really, the value you’re getting out of the house is living in it. So the price-to-rent ratio is lower in Houston than in San Francisco. And so observing prices going up faster in San Francisco and observing that the price-to-rent ratio is high doesn’t mean that San Francisco is too expensive and that prices will go back to the level they’ve been in Houston. Those two cities have been growing further and further apart for a long, long time, and I expect that to continue.
Do you see the same pattern in New York that you see in San Francisco?
Yes, similar. Real estate values in New York have been growing faster than in the rest of the country since a long time before World War II. The big difference in New York was that in the 1970s, the city really went through serious troubles and was facing bankruptcy. So I think that’s the difference between New York and San Francisco over a longer horizon. Today, particularly if you think of Manhattan, it’s just been an incredibly strong run for I think a long time, and I expect Manhattan to continue to be a place where prices rise. Manhattan is not in our study because we’re looking at single-family houses.
Why do house prices appear to be out of sync with the annual cost of owning a home?
Housing markets aren’t perfectly efficient, and there are other factors that go into driving house prices beyond just the annual cost of owning. A big thing has to do with down payments. One of the reasons we see prices in booms start to run up more quickly is that when house prices go up, people who own a home get a big capital gain. That gives them money that they can use as a down payment to buy their next house. So you see a lot of trade-up activity as prices start rising, and that fuels the number of transactions, and it also fuels some amount of price appreciation as well. So there’s a bit of a self-fulfilling aspect to prices rising, not because people have unrealistic expectations of the future, which is what a bubble is, but just because they have more cash in their pockets from the appreciation of housing to date.
Can you comment on where you think the housing market is headed?
I think we’re starting to see helpful signs of the housing market slowing down. I don’t believe that there is a bubble in housing most people live in, which is single-family houses in the United States. I think there are very few markets where people should say, “I’m paying prices that are so high that I just shouldn’t be buying a house right now.”
I do think there is a bubble in condominium markets in some parts of the country. It’s really important to emphasize that the fundamentals of the single-family housing market are different than for condominiums. Other than in the few supply-constrained markets like Boston and San Francisco and New York, I tend to think of condominiums as not being a market where you can expect a lot of appreciation. I’m not at all optimistic about where condominium prices are going to go in the next five years. I’m not sure it’s going to be very pleasant to be somebody owning condominiums in South Florida. Southern California to some extent is also problematic. The big thing with condominiums is you can build as many of them as you want. Every time the price goes up, people build more, and those markets are just getting overbuilt in an appreciably serious way.
In the housing market in general, we’re looking for a slowdown in transactions. I think we’re going to see price appreciation slow markedly to much more normal levels. And if long-term rates interest rise, I do think we will see house prices fall — but again, not because there’s a bubble, but because fundamentals have changed.
Chris Mayer is the Paul Milstein Professor of Real Estate and director of the Paul Milstein Center for Real Estate at Columbia Business School.
Read the Research
"Bubble Trouble? Not Likely"