In Depth Analysis: CalculatedRisk Newsletter on Real Estate (Ad Free) Read it here.

Friday, February 06, 2015

Lawler: Updated Estimates on the Size of the SF Rental Market

by Calculated Risk on 2/06/2015 03:41:00 PM

A long note from housing economist Tom Lawler:

While accurate and timely data on the size of the single-family rental market are not available, virtually all surveys suggest that the size of the single-family rental market surged from the end of 2007 through the end of last year. Below are some stats on the renter share of either the occupied single-family housing market (detached and attached, ACS and AHS) or the one-unit housing market (AHS, which includes manufactured housing, and is derived from detailed tables not shown in the press release).

While comparisons of these surveys with decennial Census results suggest that none of the surveys provide a particularly good “snapshot” of the overall US housing market, the fact that all surveys point to a sharp increase in the size of the SF rental market is quite consistent with anecdotal and other regional reports.

Renter Share of Occupied One-Unit Properties
Single Family (ex Mfg Housing)One-Unit Structures
(SF plus Mfg. Housing)
  ACSAHS  HVS
2007 (avg)15.0%14.1%Q4/200714.1%
2010 (avg)16.8%N/AQ4/201016.5%
2013 (avg)18.6%17.5%Q4/201318.1%
      Q4/201418.8%

In terms of decennial Census data, in 2010 Census eliminated the “long form,” which in previous Censuses had supplied more detailed information on (among many things) the characteristics of the occupied and vacant housing stock. As such, tables on the “standard” Census website for housing tenure (owner/renter) by units in structure are not available.

In one of the reports from Census providing estimation results from the 2010 Census Coverage Measurement program (#2010-G-02), however, Census does provide estimates of occupied units by tenure and units in structure. Below are two tables: the first shows “official” decennial Census results for total, occupied, and vacant single-family units (attached and detached) by tenure from Census 2000 and Census 2010, and the second shows “updated” single-family estimates for these two years based on post-Census research results.

Occupied and Vacant Single-Family Housing Units by Tenure (millions)
  TotalOwner-OccupiedRenter-OccupiedVacantRenter Share of
Occupied SF Units
4/1/200076.360.110.65.615.0%
4/1/201089.065.615.28.118.8%
Change12.75.54.62.53.8%
"Adjusted" Results Based on Post-Census Research
  TotalOwner-OccupiedRenter-OccupiedVacantRenter Share
of Occupied SF Units
4/1/200077.060.510.75.815.0%
4/1/201090.065.915.38.618.8%
Change13.05.44.62.83.8%


Focusing on the bottom table (which probably reflect the “most accurate” aggregate statistics), adjusted decennial results suggest that from April 2000 to April 2010 the total single-family housing stock increased by about 13.0 million units, but the number of owner-occupied SF housing units rose by just 5.4 million units; the number of renter-occupied SF housing units increased by 4.6 million units; and the number of vacant SF housing units increased by 2.8 million units. As a result, the renter share of occupied SF housing units jumped to 18.8% in 2010 from 15.0% in 2000.

Note that the Census 2010 data indicate that the renter share of the occupied SF housing stock was higher than other surveys (ACS and especially HVS) suggest.

If one to believe that the CHANGE in the renter share of the occupied SF housing stock since April 1, 2000 were similar to the change in that share as derived from the HVS, then (using other assumptions) the SF housing stock in the fourth quarter of 2014 would look something like that shown below.

Official Census Results
  TotalOwner-OccupiedRenter-OccupiedVacantRenter Share of
Occupied SF Units
4/1/200076.360.110.65.615.0%
4/1/201089.065.615.28.118.8%
Change12.75.54.62.53.8%
"Adjusted" Results Based on Post-Census Research
  TotalOwner-OccupiedRenter-OccupiedVacantRenter Share of
Occupied SF Units
4/1/200077.060.510.75.815.0%
4/1/201090.065.915.38.618.8%
Change13.05.44.62.83.8%
Q4/2014 Estimates based on Changes in HVS Results from Q2/2010 to Q4/2014
  TotalOwner-OccupiedRenter-OccupiedVacantRenter Share
of Occupied SF Units
Q4/2014 (est)91.265.617.97.721.4%
Change from 4/1/2010)1.2-0.42.5-0.72.5%
Change from 4/1/200014.95.47.22.36.4%


There are two “issues” with the numbers at the bottom of this table. First, of course, they use HVS results, which are not the most reliable. Second, in April 2014 the HVS began “phasing in” a new sample based on the Master Address File compiled during Census 2010, with the “phase-in” beginning in April 2014 and ending in July 2015. Since a comparison between decennial Census results and the HVS for 2010 indicated that the HVS was overstating homeownership rates and vacancy rates, a comparison between Q4/2014 results and earlier periods may not be “appropriate.” Specifically, as the new sample is phased in, one would expect the HVS quarterly data to show larger increases in the number of renters than could have been the case using the old sample.

Even if one used the change in the HVS renter share of the one-unit market from Q2/2010 to Q1/2014, however, the data would suggest that there was a considerable increase in number of renter-occupied single-family homes since 2010.

Here are what might be considered “best guess” estimates of the single-family housing stock (detached and attached) for 2000, 2010, and the last quarter of 2014.

Occupied and Vacant Single-Family Housing Units by Tenure (millions)
  TotalOwner-OccupiedRenter-OccupiedVacantRenter Share of
Occupied SF Units
4/1/200077.060.510.75.815.0%
4/1/201090.065.915.38.618.8%
Q4/2014 (est)91.266.017.57.721.0%

Much of the discussion on the sharp increase in the size of the SF rental market over the past several years has focused on “institutional” investor purchases of “distressed” SF properties, though it is quite clear that the size of the SF rental market was increasing well before most institutional investors entered the market.

There has been much less talk, however, about the demand for SF rentals, and specifically why so many households have decided to rent SF homes. E.g., how many householders are renting SF homes not because they “want” to rent, but because (1) mortgage credit has been “tight;” (2) the householders’ credit is “impaired,” in many cases because they lost their previous home to foreclosure and/or a short sale; and/or (3) they were consistently outbid by “investors” and now have been priced out of the market? On the other hand, how many householders are renting because they don’t want to purchase a home, either because they don’t like the “investment” prospects – especially those who believe they may want to move over those next few years, given the high costs of buying and selling a home – or because they don’t like the “hassle” of or potential but unknown costs associated with homeownership, but who want the “lifestyle” associated with owning a SF home? During much of the last decades householders wanting to live in a SF home in many markets faced extremely limited supplies of SF homes for rent. That is much less true today in many markets across the country.

Whatever the reasons, the surge in the size of the SF rental market has helped “sop up” much of the “excess building” during the first seven years of the previous decade, with the vast bulk of that “excess being in the built-for-sale SF market.

If for various reasons – e.g., an easing in mortgage credit, improved prospects for better-paying jobs and/or changes in potential mobility, shifts in home price expectations, etc. – an increasing number of current renters decided they wanted to own a home, then what might happen is (1) vacancy rates on existing SF rentals might increase and rents soften; (2) an increasing number of SF rental properties might be put up for sale; and (3) the resulting rise in the demand to purchase homes would not fully need to be met with an increase in the construction of new SF homes. (Just sumpin’ to think about.)

On the next page is a table showing the renter-occupied share of SF detached homes as measured by the ACS for 2006 and for 2013 by state. (Note: previous tables showed the renter share of SF detached and attached homes, which is higher. I just had already constructed this table.) This share increased from 2006 to 2013 in every state save for North Dakota, and the share jumped by more than three percentage points in 24 states.

Renter Share of Occupied SF Detached Homes, ACS
20062013Chg
Alabama14.7%18.1%3.4%
Alaska14.0%16.0%1.9%
Arizona14.7%22.8%8.1%
Arkansas18.6%20.6%2.0%
California18.9%24.1%5.3%
Colorado13.1%16.9%3.8%
Connecticut6.6%8.9%2.3%
Delaware7.4%10.1%2.7%
District of Columbia12.7%14.6%1.9%
Florida13.0%18.4%5.5%
Georgia15.2%20.5%5.3%
Hawaii23.4%25.7%2.3%
Idaho15.0%18.2%3.2%
Illinois9.8%13.0%3.3%
Indiana12.3%15.7%3.4%
Iowa12.1%14.2%2.0%
Kansas15.1%17.9%2.8%
Kentucky14.3%17.6%3.3%
Louisiana17.1%18.7%1.6%
Maine8.8%11.8%3.1%
Maryland8.5%10.5%2.0%
Massachusetts6.2%7.4%1.3%
Michigan9.6%14.1%4.5%
Minnesota6.2%9.0%2.8%
Mississippi17.1%20.5%3.4%
Missouri13.5%16.9%3.4%
Montana15.9%19.0%3.2%
Nebraska15.6%17.1%1.4%
Nevada17.3%27.4%10.1%
New Hampshire7.0%8.6%1.6%
New Jersey6.6%8.5%1.9%
New Mexico16.8%19.3%2.5%
New York8.9%10.4%1.5%
North Carolina17.0%20.0%3.0%
North Dakota12.7%11.1%-1.6%
Ohio11.8%16.0%4.2%
Oklahoma18.6%21.9%3.3%
Oregon15.6%20.1%4.5%
Pennsylvania8.9%10.9%2.1%
Rhode Island8.5%10.4%1.9%
South Carolina16.1%17.0%0.9%
South Dakota14.7%15.6%0.9%
Tennessee14.6%18.1%3.5%
Texas14.8%18.0%3.2%
Utah9.8%14.1%4.3%
Vermont10.0%10.3%0.3%
Virginia12.9%15.8%2.9%
Washington14.1%17.9%3.8%
West Virginia14.4%16.4%2.0%
Wisconsin8.2%11.0%2.7%
Wyoming15.4%16.4%1.0%
US13.1%16.7%3.6%


Rental Share Single Family Click on graph for larger image.