Wednesday, June 9, 2010

Interest Rates and the Housing Cycle

Copyright, The New York Times Company

The housing boom and bust left us with a sea of foreclosures and other economic troubles. Higher interest rates might have toned down the housing cycle but would have created other problems.

Houses may be the ultimate long-term investment. They last a long time — some houses are around for more than a century — and most last longer than the people who built and first lived in them. So the value of a house depends not only on how it will be used when it is first constructed, but also on the present value of its uses 20, 50 and even 100 years later.

The longevity of homes is the reason why interest rates, especially long-term rates, are an important feature of the housing market. The lower our long-term interest rates, the more weight the market puts on value created in the distant future, and the more houses are worth.

Housing certainly would have boomed less in the 2000s — prices would have been lower and housing construction would have been less — if interest rates had been higher. But economists take this conclusion too far when they blame a large fraction of the housing boom on low interest rates.

There are plenty of long-lived assets aside from owner-occupied houses, and interest rates also matter in determining their values. The Empire State Building, for example, is still generating significant rental incomes for its owners 80 years after its construction began. And many apartment buildings date from that time.

The chart below shows how the boom in housing construction was actually a bust in the building of nonresidential structures, and less of a boom in the building of apartments, even though the values of all three types of structures are affected by interest rates.

The series in the chart (from this paper and based on measurements by the Bureau of Economic Analysis) are indexes of construction activity for three types of structures: Owner-occupied housing, tenant-occupied housing (typically apartments) and nonresidential structures. An index value of, say, 110 means that construction activity was 10 percent greater than it was in the year 2000.

All three series increased before the year 2000, in part because the population and the economy were growing. Some measures of interest rates were low in the early 2000s, and that’s when housing construction boomed. But the construction of apartments boomed much less, and the construction of nonresidential structures was actually low — below 90 — through 2007.

Thus, the major factors in the housing boom were something other than low interest rates and are much debated (Paul Krugman and Raghu Rajan recently debated the role of home loan lending standards, and I have written about the role of technical change).

Had the Federal Reserve been able and willing to raise long-term interest rates in the early 2000s, that might have softened the housing cycle but would have worsened a shortage of business capital on which many of our jobs depend.

No comments: