The most interesting comment to my last post focused on one narrow issue: to what extent are vacant housing units second homes (and thus presumably less likely to be rented out) as opposed to units for rent/sale or held for other unknown reasons?

Why does this matter?  Because one might argue that even if overall vacancy rates are low, a high “second home rate” might be evidence that the city’s Housing Prices are rising because of nonresident investors.

Unless I am missing something, 2015 American Community Survey data does not contain data at this level of detail.  However, 2000 and 2010 Census data contains data on types of vacancies.  Below are percentages of vacant units held (in the Census Bureau’s words) “for seasonal, recreational or occupational use.”

2000                2010

Expensive markets

Manhattan                   32.7                 33.9 (3.3 pct of all housing units)

San Francisco              22.4                 17.9

Boston                         12.6                 15.2

Los Angeles                7.8                   7.9

San Diego                   26.8                 23.5

Not-so-expensive markets

Dallas                          4.6                   3.7

Houston                      6.5                   4.5

Philadelphia                2.5                   3.2

Chicago                       5.0                   7.0

On the one hand, expensive markets tend to have more second homes (evidence of a wave of outsider capital).

But if such outsider capital was a major cause of rising housing prices, one would expect the “second home percentage” to grow over the 2000s.  Instead, this number has been pretty stable.  Moreover, this group of vacancies is a pretty small percentage of the overall housing market- a bit over percent in Manhattan, and a little over 1 percent in New York City as a whole (since second homes are not so common in the other boroughs).