The U.S. housing vacancy rate--an unemployment rate for home--is at its highest level since at least 1965 (see figure). Why? Is it sticky prices? Lack of aggregate demand? Structural?
For labor, the recalculation story says that employment is a by-product of patterns of specialization and trade. Old patterns are constantly becoming unsustainable, and new patterns are constantly being created. When old patterns become unsustainable faster than new patterns are created, we have higher unemployment. Important trends over time include the increased specialization of the labor force and the shift away from labor as a variable input to production and instead toward labor as an organizational capital input.
The allocation of land and of housing units is also a by-product of patterns of specialization and trade. Technology affected these patterns. Riverboats, trains, and automobiles altered patterns of land use. The Internet may do so as well.
The Stevenson-Wolfers theory is that since 1970 marriage has been driven more by consumption complementarity than by production complementarity. I think that one can see a similar trend in location. I know of many young women who want to move to New York City because of the way they would like to live, not because of the work opportunities it provides.
Suppose that the choice of housing and location nowadays is based less on the production complementarity of being close to particular work zones (more people can work from home or from satellite offices, thanks to the Internet) and more on the consumption characteristics of the housing unit. In that case:
1. Because people have different consumption preferences (including preferences related to locational amenities), housing will become more specialized. The problem of matching housing characteristics with consumer preferences will be more challenging, just as the problem of matching worker skills with occupational demands has become more challenging.
2. If people care relatively more about consumption complementarity and relatively less about work complementarity in their choice of location, then the correlation between housing vacancies and unemployment may be reduced.
3. People may be less eager to obtain inexpensive housing, because moving to a location far from one's preferred amenities is perceived as a big sacrifice. This may slow the process of adjustment in the housing market.
So, when you overbuild houses in Nevada or condos in Florida, you cannot lure very many people with lower prices. Most unoccupied houses have close to zero marginal value to the vast majority of consumers, just as most unemployed workers have zero marginal product to the vast majority of firms. A long, difficult adjustment process is required before unoccupied houses can be matched with people who want to live in them, just as a long, difficult adjustment process is required before unemployed workers can be matched with firms that can put them to productive use.