MND recently summarized an analysis by the Urban Institute about a recent Fannie Mae pilot program to underwrite securitization of a $1 billion loan written by Wells Fargo to Blackstone Group, collateralized by thousands of single-family rental properties.  That article alluded to the large number of formerly owner-occupied homes that became rentals (an estimated three million) during the housing crisis.

The Mortgage Bankers Association (MBA) has just released a study from the Research Institute for Housing America (RIHA), authored by Stuart S. Rosenthal, Maxwell Advisory Board Professor of Economics at Syracuse University, that indicates that such a shift was not necessarily an unusual one.

 "A striking feature of the last housing crisis was the dramatic shift of owner-occupied homes into the rental sector.  This paper looks back historically to help understand how common such shifts have been and finds that the factors affecting the rate of change are quite similar over time," said Lynn Fisher, Executive Director of RIHA and the Mortgage Bankers Association's Vice President of Research and Economics.  "Also, it's important to understand the dynamics of the existing housing stock, because they help explain how the market provides affordable housing, and how it adjusts in the face of changes in both supply and demand.  This is especially true today, when new homes are adding such a small fraction to the existing home stock."

Using longitudinal data from the U.S. Census Bureau's American Community Surveys covering 2000-2014 and the 1985-2013 American Housing Survey panel it found several patterns involved in the occupancy shifts.

First, it found that, over time, transitions occur not only from owner-occupancy to rentals but in the opposite direction as well. On average, each decade there is a net transition of about 2 percent of the housing sector into rentals.  However, the majority of transitions are to rentals as homes age.

The second finding was that, in the short-term, homes frequently transition driven by home prices.  Rising prices draw rental units into owner-occupancy while falling prices have the opposite effect. 

There is also data that indicate a shift to a rental property is more common when an owner-occupant is under water, that is, owes more on the mortgage than the value of the home.  For homes only slightly underwater, a combined loan-to-value (CLTV) ration between 100 and 120 percent, the probability of becoming a rental property is only 1 to 2 percentage points greater than a home that isn't in negative equity.  Homes that are deeply underwater, with a CLTV exceeding 120 percent, the owned-to rented probability is 6 to 8 points higher than for homes with equity.

Further analysis reveals that these transitions occur primarily for housing types for which there is ample demand in the rental market. For underwater properties that have characteristics that limit demand in the rental sector, transitions to rental status largely do not occur. Moreover, these patterns were nearly identical pre- and post-financial crisis which suggests that the mechanisms that govern the impact of high CLTV on housing stock transitions were similar in both periods. These data also make clear that the availability of mortgage financing affects the potential for own-to-rent housing stock transitions in the face of falling housing prices.

Rosenthal sees a pattern of both short-term and long-term transitions.  Short run transitions are sensitive to changes in market conditions such as rising house prices that may affect anticipated returns to investment, or falling house prices that may trigger defaults.  These transitions are likely to be reversed along with reversals in market conditions, for example, a rebound in home prices.  Long run transitions are more likely to be sensitive to age-related depreciation of the housing stock and are part of a filtering process by which markets produce affordable housing.  Both long and short term transitions are more frequent for housing types with a ready source of demand in the opposite sector and therefore are also sensitive to structural and neighborhood attributes of a home that affect long and short run perceptions of housing quality.

Rosenthal says his findings have several implications for housing market dynamics including the long-run availability of affordable housing.  Homes in the rental segment tend to filter down to households of lower real income at a rate of about 2.5 percent per year as compared to 0.5 percent for homes in the owner-occupied segment. Transition of housing stock into the rental sector, therefore, amplifies this filtering process and accelerates the rate at which markets provide affordable housing.

Another implication is for residential construction.  The author says movement from rental to owned housing has the potential to undercut demand for new construction.  In addition, the current low national homeownership rate suggests that a large buffer stock of potential owner-occupied homes may now sit in the rental segment of the market.  He offers this as one explanation for why new home construction remains so far below expectations even as home prices have returned to pre-crash levels and says additional research into this relationship is warranted.