by Calculated Risk on 2/11/2008 12:33:00 PM

The usual real estate refrain is location, location, location. But right now, inventory is the key to understanding the housing market.

As I noted yesterday in Housing as an Engine of Recovery, housing usually leads the economy both into and out of recessions. But for this year I argued:

... given the current fundamentals of housing – significant oversupply, falling demand – it is very unlikely that housing will act as an engine of growth any time soon. We need to see a significant reduction in supply before there will be any increase in residential investment.

So, for those expecting a 2nd half recovery in the economy, I believe they need to look elsewhere for growth – and they need to argue this time is different, i.e. that the economy will recover before housing ...

So let's look at inventory, but first a funny quote:

"[T]he homebuilders have basically stopped building -- they are building one-quarter of the homes they did in 2006 -- we are going to run out this year."
Jim Cramer, 01/31/2008

That is factually wrong on every point. The homebuilders have not stopped building, they are building far more than one-quarter of the homes they built in 2006, and - most importantly - the housing market is not going to run out of inventory any time soon.

There are several different ways to look at housing inventory. The most frequently mentioned measures are new and existing home inventory levels released monthly by the Census Bureau and National Association of Realtors (NAR) respectively.

The first graph shows new home inventory in December - houses for sale, seasonally adjusted (SA) – from the Census Bureau.

Click on Graph for larger image.

The 495,000 units of inventory for sale at the end of December is slightly below the levels of last year.

At first glance it appears new home inventory is declining. However there are a couple of important issues with new home inventory. First, the Census Bureau ignores cancellations (here is the Census Bureau description of how they handle cancellations), so during periods of rising cancellation rates, the Census Bureau overstates New Home sales and understates the increase in inventory. Conversely, during periods of declining cancellation rates, the Census Bureau understates sales. Second, new home inventory excludes many condominiums, and in certain communities (like Miami and San Diego) there are anecdotal stories of a glut of condos.

By my calculations, based on cancellations, the inventory of new homes is currently understated by about 100K. Unfortunately there is no available data source to adjust for excess condos.

New home inventory is just a small part of the picture. The next graph shows nationwide inventory for existing homes. Note: Unlike the new home inventory data, the existing home inventory data is not seasonally adjusted.

According to NAR, inventory was down slightly at 3.905 million homes for sale in December.

Total housing inventory fell 7.4 percent at the end of December to 3.91 million existing homes available for sale, which represents a 9.6-month supply at the current sales pace, down from a 10.1-month supply in November. “The fall in inventory in December is encouraging, but inventories remain elevated and buyers have a clear edge over sellers in many markets,” Yun said.

The typical seasonal pattern is for existing home inventory to decline sharply in December (usually by about 13%), as homeowners take their homes off the market for the holidays. So, not only is this the highest yearend inventory in history, but the December decline was less than normal (only 7.4%).

Another way to look at excess supply is to use the homeowner and rental vacancy rates from the Census Bureau.


The third graph shows the homeowner vacancy rate since 1956. The current rate is 2.8%, well above the recent normal rate of about 1.7%. There is some noise in the series, quarter to quarter, but it does appear the vacancy rate has stabilized.

This leaves the homeowner vacancy rate about 1.1% above normal, or about 825 thousand excess homes.

Sometimes rental units are a reasonable substitute good for single family homes. So we also need to consider the rental vacancy rate, and calculate the excess rental units.

The rental vacancy rate declined to 9.6% in Q4, from 9.8% in Q3. The rental vacancy rate has been trending down slightly for almost 3 years (with some noise). This was due to a decline in the total number of rental units in 2004, and more recently due to more households choosing renting over owning.



It's hard to define a "normal" rental vacancy rate based on the historical series, but we can probably expect the rate to trend back towards 8%. This would suggest there are about 560 thousand excess rental units in the U.S. to be absorbed.



Here is a rough estimate of the excess inventory:

Source	Units
Rental Units	560,000(1)
Vacant Homeowner Units	825,000(2)
Excess Builder Inventory	250,000(3)
Total	1,635,000

(1) According to the Census Bureau there are 35.12 million rental units in the U.S. If the rental vacancy rate declined from 9.6% to 8%, there would be 1.6% X 35.12 million units or about 560,000 units absorbed.
(2) Based on the homeowner vacancy rate declining from 2.8% to 1.7% on 75 million units.
(3) Based on a return to 5 months of hard inventory (completed or in process). 100,000 additional units are included based on rising cancellation rates.

Until the level of inventory declines significantly, housing prices will continue to decline and the outlook for new residential investment will remain grim.