The Census Bureau reported that new homes sales fell in May to a seasonally adjusted annual rate of 342,000, a 0.6% decline from Aprils 344,00) rate. However, the April number was revised down from an initial read of 352,000.

The May rate was a fairly significant disappointment relative to expectations of a 360,000 annual rate. On a year-over-year basis, sales are down 32.8%.

There was some good news in the report. For starters, all of the weakness was concentrated in one region, namely the South, where sales were off 8.5% for the month, and 35.9% versus May 2008. The South is by far the biggest and most important of the four census regions when it comes to housing. Even with the decline, it represented 53.8% of all new home sales.

That amounts to a sharp decline from the 58.4% share in April. This is because the other three regions all posted increases on a month-to-month basis.

Sales were especially robust in the Northeast, where they rose 28.6%. Still, the Northeast is tiny when it comes to housing, and even with the increase it accounted for only 7.9% of all new home sales, up from 6.1% in April. The Midwest posted a robust 18.6% increase for the month, but is still off 32.0% on a year-over-year basis. The West saw a 1.3% increase for the month, and a 31.0% decline on a year-over-year basis.

The other good news was on the inventory front. The stock of new houses waiting to be sold fell 2.3% on the month and is down 35.5% on a year-over-year basis. This caused the months supply figure to fall to 10.2 months from 10.4 months in April and 10.7 months a year ago. It also represents a substantial decline from the 12.4 month peak set back in January.

The absolute number of houses for sale has been down every month for the last year. Slowly but surely, we are working off the inventory overhang in new housing.

The two graphs below (both from http://www.calculatedriskblog.com/) illustrate what is going on with regard to inventories. The first shows the months of supply metric. It shows that while we are well off the record peak, we are still at levels that are just plain awful from a historical perspective. While the four-month level that prevailed in the late 1990’s and early 2000’s might have been an aberration, we would still have to get down to about six month supply for the market to be considered anything near healthy.

On the other hand, months supply is a function of both inventories and of sales, and sales are near their lowest level since the early 1960’s (well, technically January’s 329,000 rate was the record low, and we are less than 4.0% above that). The absolute level of inventories, shown in the second graph, is back to historically normal levels. Thus if sales were to pick up, the months of supply number could fall in a big hurry.

I don’t see that happening yet, though. Existing homes make a pretty good substitute for new homes, and there is still a huge supply of those available at distressed price levels. People who are afraid they are going to lose their job don’t rush out to buy new houses. Instead, I see a steady chipping away at the inventory figures as new home sales stabilize around these levels for the rest of the year, and reduced levels of housing starts choke off the river of homes flowing into the inventory pool.

Eventually, when the economy gets back on its feet, the housing industry will revive. The housing sales numbers (and inventory) are not adjusted for population growth, and over time new household formation will cause pent-up demand for housing. Kids who have moved back in with Mom and Dad after college will start to chafe, and Mom and Dad will get fed up. People bunking on friend’s couches because they have been foreclosed on will get back on their feet financially and move back out to their own places.

However, there is a serious chicken-and-the-egg problem here, since historically residential investment (which is for the most part new housing activity) is the locomotive that drives the U.S. out of recessions. That locomotive is derailed — meaning the recovery is going to be extremely slow and anemic.

From a GDP growth point of view, the absence of a negative will be a positive, but not as much as if it were actually a positive contributing to growth. I’m not ready to recommend the homebuilders like D.R. Horton (DHI), Beazer (BZH) or Centex (CTX), but I don’t think I would be shorting them at this point either.

Read the full analyst report on “DHI”
Read the full analyst report on “BZH”
Read the full analyst report on “CTX”
Zacks Investment Research

Last 5 posts by Dirk Van Dijk

• Exponential Growth, Finite World - Analyst Blog - November 20th, 2009
• Give Thanks for a Great 3Q - Earnings Preview - November 20th, 2009
• Mortgage Delinquencies: Record High - Analyst Blog - November 19th, 2009
• Initial Jobless Claims Flat(-ish) - Analyst Blog - November 19th, 2009
• Zacks Industry Rank Analysis Highlights: B.P. Prudhoe Bay Trust, San Juan Basin, Exxon Mobil and Petrobras - Press Releases - November 19th, 2009