While existing home sales have recovered to pre-bubble 2003 levels, new home sales remain about a third below the levels seen any time in the past fifteen years.
In the years 1994 through 2006 (and probably well before then), new home sales were consistently about one-sixth of existing home sales. The ratio of existing home sales to new home sales has risen since then as the housing bubble unraveled. Right now, new home sales are one-fourteenth of existing home sales.
This means that it may take a very long time for the home construction industry to recover. First, existing housing inventories need to be absorbed. The ratio has been rising for almost four years, and even if it returns to normal levels in half that time, this means that the new home construction industry would not return to pre-housing bubble burst levels until early 2012. It could take longer.
Construction and real estate are quintessentially local industries, and the housing bubble was not a national phenomena. In markets like California, Arizona, Nevada and Florida there was an intense housing bubble and subsequent collapse, so there is a good chance that those markets are very overbuilt and that the new housing industry recovery there will take longer. The Denver market and many markets in the American heartland did not see nearly as much of a housing bubble, so there is a good chance that those markets are not very overbuilt and that the new housing industry can recover more quickly there.
This is complicated by the fact that the manufacturing industry has taken a big and probably partially permanent blow in the current financial crisis. The automotive industry through the bankruptcies of General Motors and Chrysler, finally carried out a downsizing that had been looming for years as these companies lost market share. The "New Rust Belt" that runs through the American South and consists of counties with heavily manufacturing reliant economies in otherwise relatively rural areas may have taken a blow as well, although there is reason to think that these newer ventures are more flexible and more prone to come back as the economy recovers. Thus, at least in the Old Rust Belt, the new housing market may be hit just as hard as it was in areas that had severe housing bubbles, and may not come back anytime soon.
Short term stimulus packages are unlikely to provide a long term benefit to the new housing industry in either busted housing bubble areas or areas that have seen permanent declines in their manufacturing job base in the Old Rust Belt. Incentives to buy homes may eat away at the excessive housing inventory for a while, but those incentives also reduce demand in the months that follow because potential new home buyers already bought homes a little bit sooner than they would have in the absence of incentives. When the excess housing inventory is small, this kind of stimulus can return the economy to normal. But, when we have a huge excess inventory, as we appear to now, the excess inventory is likely to outlast the stimulus, so there is little benefit in the medium or long term.
Since the construction industry, particularly in the area of new housing, was a major driver of immigration, this may also mean that immigration, legal and illegal, will recover more slowly than the economy as a whole. This takes pressure off policy makers, as immigration becomes a less significant issue in the economy and also provides a window in which the kind of immigration reforms that President Obama ran for office upon may open up.