Is the Long-Term Trend in Housing Prices Up? Or Down?

BusinessWeek's cover story in its February 11, 2008, issue addresses the recent boom--and even more recent bust--in housing prices. A chart on p. 41 suggests that, leaving out the past 8 - 10 years, housing prices show no trend at all over the 1890 - 2000 period. BusinessWeek calculates average annual rate of growth in housing prices, adjusted for inflation, at 0.4% per year since 1890 (using Robert Schiller's index of housing prices)--including the 2000 - 2006 boom years.

I suspect this will strike many people as hard to believe. After all, many people's single most valuable asset is the home that they occupy, and one of the reasons often advanced for home-ownership is the rising value of that home over time. Yet the long-term data suggest that there is little or no rising home value.

What's going on?

The answer, oddly enough, is fairly straight-forward. There is a large amount of research (here, here, here, here) suggesting that housing construction is extremely sensitive to changes in the price of housing. In economic terms, we would say that the price elasticity of supply of housing is large. How sensitive? Well, the estimates suggest that a 1% permanent increase in the price of housing would lead to an increase of between 3.7% and 13% in the number of housing units. In short, the extraordinary responsiveness of home-building to price increases--and the equally extraordinary response to price decreases--essentially holds the long-term price of housing very near its long-term averag level.

Housing prices begin to spike, and construction follows, just as it did between 2000 and 2006. Housing prices begin to slide, and construction follows, just as it has since 2006.

The conclusion, of course, is that the extraordinary boom in housing prices in the past few years has led to a degree of over-shooting in housing values that is literally unprecedented in the US over the past century. So the spreading belief that housing prices will undergo a fairly long--and for recent home-buyers, painful--adjustment simply says that we're going to return to the long-term expected values for housing.

It also suggests that research by Gregory Mankiw and David Weil, mentioned in the BusinessWeek article, predicting a 47% inflation-adjusted decline in housing prices between 1989 and 2025 is also somewhat off the mark.

The real adjustment will come in the size of the housing stock, although short-term housing price fluctuations may continue to make people think that the golden era of home ownership has arrived--or that all is lost.

Neither is true.