by Calculated Risk on 9/05/2012 02:31:00 PM

CR Note: Housing economist Tom Lawler estimates that there are about 2.1 million more single family home rented now than in 2006. This is a key reason for the decline in inventory.

From housing economist Tom Lawler:

Recently there have been a sizable number of media stories on the SF rental market, with the focus on its tremendous growth over the past few years. One reason for the jump in the number of articles is related to the significant increase in the number of entities who have entered this space. It seems almost as if everyone and their mother has either entered the SF rental market or is looking to enter it. Another is that investor demand for SF properties has been very strong while the supply of homes for sale, especially the supply of foreclosed homes, is down significantly, and as a result prices of “distressed” (and other) homes have increased significantly faster than many had expected. And finally, of course, there are stories that a “REO-to-Rental” securitization deal is in the works (if it comes it’ll probably be unrated, as (1) rating agencies don’t really have sufficient data to assign a rating; and (2) unlike in the past, currently rating agencies care about such things!)

Stories of the surge in the SF rental market are not, by the way, limited to California, Arizona, Nevada, or Florida, but are reasonably widespread across the country.

Of course, the explosion in the size of the SF rental market is not new: it was evident several years back. But the growing number of “new” players, combined with the sharply lower inventory of homes for sale and recent rebound in home prices, has made this an increasingly important “story” for the housing market.

It is, unfortunately, not easy to get a good handle on just how rapidly the SF rental market has grown, given the lack of good, timely information on the US housing market. Data from the American Community Survey, e.g., are only available for 2010, and there are some “issues” with that data (as evidenced by the ACS/decennial Census differences currently being explored by Census analysts). There are even bigger issues with data from the Housing Vacancy Survey, which deviated incredibly from the decennial Census (and ACS) on a wide range of “metrics,” and whose estimates appear to systematically understate the number of renter-occupied households. Moreover, the HVS does not explicitly release estimates of the number (or %) of owner vs. renter occupied homes by units in structure.

The HVS does, however, release estimates of (1) the rental and homeowner vacancy rates by units in structure; (2) the % of vacant homes that are 1-unit structures; and (3) the number of total homes for rent and for sale. While unfortunately a consistent time series of these estimates doesn’t go back very far, and unfortunately the HVS relies on the American Housing Survey for estimates of the characteristics of housing units (the AHS doesn’t come close to matching the ACS), the data may have some useful information on trends in the SF rental market.

Unfortunately (gosh, I use that adverb often when describing available US housing data), the HVS’ definition of “1-unit” structures includes not just SF detached and attached homes, but also mobile homes or trailers, tents, and boats.

With that in mind, here are some data on the share of occupied “SF” homes that were occupied by renters from (1) the American Community Survey, and (2) derived shares using the aforementioned tables from the Housing Vacancy Survey. The latter are for the second quarter of each year, as the tables released are quarterly, and it’s a pain to derive yearly average data. The ACS data are based on the one-year estimates. Also shown are comparable “SF” shares from Census 2000.

	Renter Share of Occupied "SF" Homes
	American Community Survey*			Housing Vacancy Survey**
	SFD	SFD+SFA	SFD+SFA+ MH+Other	SFD+SFA+ MH+Other
Census 2000	13.2%	15.0%	15.6%
2006	13.1%	14.8%	15.7%	14.5%
2007	13.4%	15.0%	15.9%	14.7%
2008	14.0%	15.7%	16.5%	15.1%
2009	14.8%	16.5%	17.3%	15.4%
2010	15.1%	16.8%	17.6%	16.3%
2011			18.0%	16.7%
2012			18.3%	17.0%
SFD - Single family Detached SFA - Single family Attached MH - Manufactured Housing Other - Boats,RVs, Vans, Tents, Etc. * Yearly Average ** Q2 Average

As the table indicates, both surveys suggest that the renter share of the SF market has increased significantly since the beginning of the housing bust. Given the systematic tendency for the HVS to understate the renter share of the overall housing market (as well as the number of renter-occupied homes), it is not surprising that the HVS estimates (again, derived from table not in the press release) of the renter share of occupied “one-unit” homes is below that of the ACS, though the differential between the two hasn’t changed radically over time.

On September 20th Census plans to release the 2011 ACS results, and the above data strongly suggest that the rental share of the SF market increased from 2010 – and the 2012 ACS data will almost certainly show a gain form 2011. A “reasonable” best guess, based on the admittedly “iffy” HVS data, would be that the 2012 ACS data will show that the renter share of occupied SF detached homes this year will be about 15.7%.

Assuming ACS data were correct, such a share increase would imply that the number of renter-occupied SF detached homes in the US this year is about 11.4 million, almost 2.1 million (or 22%) higher than in 2006, with most of that increase coming after 2007. Not coincidentally, foreclosures ramped up sharply in the latter part of 2007, and REO sales increased significantly in 2008 and remained high through last year.

Renter Share of Occupied SF Detached Homes (ACS-based)

Click on graph for larger image.

These data are broadly consistent both with anecdotal evidence and by statements from some of the “larger” players in the SF investor space that a fairly large % of investors buying “distressed” SF properties have purchased the home with the intent to rent the home for “several” years – partly because in many parts of the country distressed home prices were low relative to realizable rents (in other words, the “rental yield” was good), and partly because investors expected home prices several years down the road would be higher than current prices.

The surge in the number of properties purchased with the intent to rent (at least for a while) has also almost certainly contributed to the sharp decline in the number of homes listed for sale.