The House Price Bubble: Won't Get Fooled Again

Despite widespread talk, the evidence for a housing bubble is still murky. Here's where we should expect problems.

Article created by The Century Foundation.

One of my all time favorite rock albums is The Who’s “Who’s Next” and one of my favorite tracks on that album is “Won’t Get Fooled Again.” Right now there is much talk of a housing bubble, making for the possibility that a lot of people are getting fooled.

There are two ways to assess whether there is a housing bubble. The first can be termed the “historical approach,” and it involves looking at the historical relationship between house prices, levels of income, interest rates, and demographic factors. According to that approach house prices look significantly out of whack.

The second can be termed the “comparable cost approach,” and it involves comparing the relative cost of renting versus buying. A recent study by Professors Gary and Margaret Smith of Pomona College in Southern California uses this approach and concludes that house prices are generally not over-valued. Their findings have quickly been advertised by the mainstream media (New York Times, Saturday April 1, 2006), but the comparable cost approach has serious limitations. In the spirit of openness, here are some cautions about their conclusion.

People buy houses to enjoy the accommodation services that houses provide. An alternative way to get those services is to rent an equivalent house. If over the lifetime of occupancy, a house can be rented for less than it can be purchased this would suggest house prices are over-valued. Such a comparison involves a complicated calculation involving assumptions about future rents, future home ownership costs, and future house price appreciation that determines what an owner gets when they sell. Using this method, and assuming future annual rent increases and house price appreciation of three percent, the Smiths found that house prices were fair to under-valued, except in a few areas.

The main problem with the Smith’s study is that it assumes away the bubble. Their baseline calculation assumes a three percent annual increase in house prices, which automatically means no bubble. If prices keep rising, there cannot by definition have been a bubble.

That leads to the core problem with the comparable cost approach, which is that it provides no insights into future prices of houses or rents. Instead, it compares existing house prices with existing rents and factors in an assumed future path for prices and rents. That makes it a good tool for framing the “rent or buy” decision, but not for predicting future prices. It is possible that rents today are too high and could fall and drag down house prices. Alternatively, house prices could fall and drag down rents. There is some evidence that both may happen, but this vital evidence is ignored.

House builders are reporting record profits, which means that cost of building homes is significantly below the price of homes. Consequently, builders have an incentive to keep building homes and adding to the supply of homes until prices return closer to building costs. Given today’s prices and costs, new house construction promises to keep the lid on home prices and possibly lower them.

The same holds for the rental market. Real Estate Investment Trusts have also reported record profits. This suggests that rental properties are profitable, providing an incentive to build more rental units, which will keep the lid on rents. This incentive is strengthened by the fact that rental supply has been reduced by conversion of rental units into condominiums.

Either of these effects—flat home prices or flat rents—dramatically changes the conclusion that houses are fairly valued under the comparable cost approach. Note, that neither prices nor rents need to fall to generate the conclusion that people in many locales are over-paying. All that is needed is for prices or rents to flat-line, in which event buying a house is a poor investment at today’s prices.

A second core problem with the comparable cost approach is that it completely ignores risk. For renters, the main risk is faster than expected future rental price increases. For buyers, the main risk is capital loss, and this risk is closely linked to how long one holds the house. If you hold a house for thirty years, it is unlikely you will incur a capital loss because the long-run trend of house price appreciation will dominate temporary price fluctuations. However, if you hold it for less than ten years, the likelihood of loss is much larger—particularly after a period of rapid home price increases.

Given the leveraged nature of home buying, small decreases in home prices can inflict large losses on owners. Consider a $500,000 home with an initial ten percent down payment. If the house price falls ten percent, the owner loses one hundred percent of their equity. If the price falls twenty percent, they end up with negative equity of $50,000 that would probably bankrupt most households. This can have lifelong consequences by denying future access to credit, which can also prevent future home purchases.

These considerations are especially important to young buyers who often buy condominiums. The holding period for young buyers is shorter, exposing them to greatly enhanced risk of loss. Moreover, home prices do not even have to fall to impose large losses. Remember, the cost of selling a home is six percent. For the above $500,000 example, the exit cost is $30,000 and that wipes out more than half of the initial equity investment. The important point is that houses are illiquid and risky investment. Such investments usually trade at a discount to flexible liquid investments (i.e. renting), and homes (especially condominiums) should therefore sell at a slight discount to renting.

If there is a bubble, it is likely heavily concentrated in the condominium market. However, the effects will ramify throughout the residential real estate market. Many condo owners could find their equity wiped out, and condos are often the first step on the housing market ladder. Consequently, condo market weakness will ripple into broader housing market weakness. Over the long run prices will recover, but as the great British economist John Maynard Keynes remarked, “In the long run we’re all dead.”