It’s taken as a given that one of the reasons housing is in crisis is that home value increases have significantly outstripped income growth (see, for example, these posts at The Big Picture, Option Armageddon, and Calculated Risk). Here’s a chart from Calculated Risk showing the relationship over time:
(Click on image for a larger version in a new window)
An excerpt from Option Armageddon explains:
Ask yourself, what is a housing “bubble” and how is one created? The term “bubble” suggests that prices were, objectively speaking, “too high.” Clearly this was the case. A chart of house prices relative to median income makes it abundantly clear. House prices can’t continue to expand forever, not unless incomes expand at the same time. If prices are expanding faster than income, then prices are “too high” relative to what people can actually afford to pay for shelter. In other words, we have a bubble.
This is common sense. But is it true? If it is, you would expect that there would be more foreclosures in markets where the ratio was higher. But that’s not necessarily the case.
Via Creative Class, a study from University of Virginia researchers found:
In San Francisco, for example, median value of owner-occupied housing in 2007 was 9.7 times median family income, yet the foreclosure rate was a mere 0.24 percent. In the District of Columbia, housing values were 6.8 times family income, yet the foreclosure rate was 0.12 percent. And in New York City, housing values were 12.3 times family incomes in Brooklyn (foreclosure rate 0.38), 11.7 times income in Manhattan (foreclosure rate 0.04 percent), and 10.3 times family income in the Bronx (foreclosure rate 0.28 percent). Other central cities lacked such extraordinary house value to income ratios, but in no instance were low foreclosure rates associated with low house value to income ratios (Table 4).
Here’s the table:
If the relationship is true, why does San Francisco, which has a value-to-income ratio triple the national average, have a foreclosure rate that is 1/3 the national average?
There is clearly something going on that can’t be expressed in a simple ratio. My suggestion is that bubble markets tend to have relatively low income levels and relatively high concentrations of single family rentals (see this post for a more detailed explanation).
|