Revisiting Home Prices
Speaking as a charter member of the No Housing Bubble in America Club, we believe the time has come to take a slightly revised position. While the headline remains unchanged – there is no housing bubble in America – there is a growing risk of “bubblettes” in certain places.
The essence of our long-held no-bubble position is that low interest rates have worked wonders to enhance affordability of home ownership. So despite today’s median home price being stratospherically high relative to the median household income, low mortgage interest rates have kept payments on those homes historically low. Last year, for example, 22 percent of the median household income was required to make payments on the median priced single-family home. That figure has been largely unchanged for a decade and remains well below the 35 percent average of the 1980s. In fact, even a one point rise in mortgage rats would bring that affordability measure to only 25 percent of household income, equal to the average of the past quarter century.
Nonetheless, while median metrics do a fine job of describing the over-all housing market, there are sizable differences from place to place. During the past year, for example, house prices soared 36 percent in Nevada while advancing just four percent in Texas – a difference that seems at odds with underlying fundamentals. To explore the validity of these increasingly disparate experiences, we asked the following question: What should home prices be in each of the 99 largest metro areas after controlling for differences in population density, relative income levels, interest rates and historically observed premiums or discounts in those markets? By comparing actual prices to these norms, it is possible to identify markets that are over or under-valued.
What we find is a broad dispersion of property market valuations, ranging from a seemingly tenuous over-valuation of 43 percent in Chico, California, to an alluring under-valuation of 23 percent in Salt Lake City, Utah. To be sure, over-valuation is not pervasive, however. Markets demonstrating over-valuation of 20 percent, or more, collectively account for just about one-fifth of the U.S. housing stock. That’s not a trivial share, but it is far from a majority.
Lastly, it should be recognized that while over-valuation presents the risk of subsequent price declines, that risk may well go unrealized. To understand why, consider two stocks – Dell and Lucent. At the peak of the hi-tech equity bubble, five years ago, each had an astronomical price-to-earnings multiple in the vicinity of 100. Dell’s stock price today is unchanged from that time, however, as years of double-digit earnings growth essentially ratified lofty growth expectations. Lucent, alternatively, now trades a just one-twentieth of its once high price as steep losses have characterized much of its past five years.
So as we consider these over-valued markets, the relevant question is this: Will subsequent growth in those areas justify today’s high prices (the Dell outcome), or will those expectations prove unrealistic (as was true of Lucent)?
