Wow! Nevermind the recessions... what a runup in sales we've had! Makes me think sales could easily fall 50% or more from current levels.

Puts the inventory numbers in perspective -- even if the absolute numbers don't increase anymore, the months of supply could still skyrocket from declining sales.

Would it be possible to calulate the months of supply of new and existing homes we would have at current numbers if sales volume "returned to the mean"?

CR, I would be very interested in seeing that analysis.

The U.S. is adding 2.68 million people per year (Census Bureau). The average is 2.4 people per housing unit - so that is 1.1 million units. BUT that includes apartments...

Could the homeownership rate be used as a scaling factor here? (e.g. 1.1 million units * .7)

Also, I think it might be productive to estimate the amount of "borrowed demand" in a similar manner, combine the two factors and assume, say, a 3 year correction following an established pattern (e.g. the Texas bust) and see what comes up.

This also might give an indication of the impact on construction jobs too.

For the overall market, though, I think it might make sense to aggregate new and existing home inventory and sales since in the long run their respective trends will probably converge, and just make a distinction where it provides additional information.

DaveL,

That said, AC, not clear on your last point?

Well, I was thinking the point of this sort of analysis is to develop a "possible outcome" for the current housing market. Yes, new home sales relate specifically to fundamental oversupply issues, but I think the real question is "what's in store for the housing market and it's participants?" I think the distinction between new homes and existing homes is only relevant to certain questions ("how are construction workers affected? or speculators?") but not broader questions ("how much downward will there be on prices?").

To put it another way, you don't hear people who lived through the Texas bust saying things like "that housing market was just a complete nightmare for 3 years... except those new homes were just flying off the market no matter how high they raised the prices!!!"

From '94 through '05, the percentage of heads of housholds reporting themselves as homeowners rose about 0.5 percentage points a year. So regardless of the exact number of households formed, the construction rate was above it by that amount, which equates to about 400k units/yr. So over those ten years about 4 million of the homes sold went to increasing the ownership rate from its historical range of 63-65% to the current level near 69%. If the ownership rate reverts to its historical range, those 4 million units are going to leave the hands of their current owners and pass to the hands of others, and the number of units to be absorbed by the market is that much larger.

Some will argue that the households currently in these units must live somewhere, so they won't become vacant and available to new households, but will at worst just move into the hands of landlord investors. This overlooks the fact that much of the growth in ownership was among young people, who will end up doubling or tripling up with roommates, or moving back in with their parents.

The crash has only just begun. We face a housing glut of unprecedented scale.