Shortage of Apartments Will Not Create a Building Boom

Written by John Lounsbury

Today (April 6) GEI News has an article reporting that apartment vacancies are near an all-time low.  With all the talk about a housing glut this may come as a surprise.  It’s worth a trip through the numbers to see just what is going on.    Steven Hansen has recently reported that multifamily housing starts have seen a strong increase in recent months.  Could it be that the long drought in residential construction could be over as builders rush to meet a growing need for rental properties?  This analyst thinks not and will go through the data to explain why.

One estimate for the total number of U.S. households in 2017 is 117 million.  (See below.)

The above estimate is based on the extrapolation of estimates using 2000 census data.  Another estimate for the number of households in 2017 can be derived from the data from the 2010 census (number of households in 2010 = 116.7 million) which produces approximately 121 million using the slope of the trend line in the 2000 census based graph above.

The following graph shows the bubble in home ownership starting in the mid-1990s.

If the home ownership rate declines to the pre-bubble level of 64%, this means that 36% of residences will be rental properties by 2017, or about 43.6 million.  This is almost 3 million larger than the 2010 Census Bureau number of 40.7 million renter occupied housing units.  This would be a good number if there was not an increase in credit impairment due to mortgage defaults.

If credit impairment reduces the number of owner occupied homes below what otherwise would have occurred, the shortfall in rental properties could be much larger.  One reasonable estimate would be to assume the worst case for credit impairment to have 10 million former home owners locked out of new mortgages.  This is approximately 2/3 the total number of foreclosures estimated for the entire bubble cycle.

That would put the worst case for home ownership rates by 2017 at 56%, or about 68 million units and occupied rental units at approximately 53 million units.  This would make the shortfall for rental units at approximately 8 million units.

Thus we have bracketed the number of additional rental units needed by 2017 as being somewhere between 3 million and 8 million units.  Some of these will come from investor converted foreclosure properties that were formerly owner-occupied.  Some of these will come from new rental construction.

The tables below summarize these estimates.

There exist almost enough excess vacancies (see graph below) above normal levels from before the bubble to satisfy the need for additional housing by 2017.

A reasonable conclusion is that very limited new housing construction is needed between now and 2017 unless the conversion from formerly owner occupied to rental properties is insufficient (increasing the need for rental unit construction) or millions of unoccupied units are demolished or otherwise become uninhabitable.

The increase in multiple family housing starts reported by Steven Hansen may not have much follow through if the conversion of the excess previously owner occupied apartments, condos and single family homes takes hold.  And that process does appear to be starting.

References and Related Articles

  • Number of Households:  US (estimate, in thousands) (The Capitalist Manifesto, 5 April 2012) This appears to be based on estimates from the 2000 census data which produces a number of households well below the estimates based on the 2010 census (next reference below).  Until this is determined for certain the use of this data for projections is considered risky.  The U.S. Census Bureau has been asked for clarification of the two data values.

2 replies on “Shortage of Apartments Will Not Create a Building Boom”

Comments are closed.