From the Associated Press this morning:

"Construction of new homes jumped in May by the largest amount in three months, an encouraging sign that the nation's deep housing recession was beginning to bottom out. The Commerce Department said Tuesday that construction of new homes and apartments jumped 17.2 percent last month to a seasonally adjusted annual rate of 532,000 units. That was better than the 500,000-unit pace that economists had expected and came after construction fell in April to a record low of 454,000 units. In another encouraging sign, applications for building permits, seen as a good indicator of future activity, rose 4 percent in May to an annual rate of 518,000 units. The better-than-expected rebound in construction was the latest sign that the prolonged slump in housing is coming to an end, which would be good news for the broader economy."

Pretty lousy analysis if you ask me. It is true that more construction will show up in GDP calculations as so-called "economic growth" but the idea that growth in housing starts is good for the housing market and means the housing recession is coming to an end is completely wrong.

In case the AP hasn't noticed, housing prices are cratering due to a supply-demand imbalance. When supply exceeds demand, prices drop (economics 101). It is widely believed (and I agree) that a bottom in housing prices (and therefore an end to the housing recession) is needed before the U.S. economy can really begin a sustainable recovery (such an event would boost consumer confidence and spending, and help the banks feel better about extending credit). In order for home prices to stabilize, we need the supply-demand picture to balance out.

How will supply and demand meet if we build more supply when the problem has been (and continues to be) an excess supply of unsold homes? They won't, which is why a pick up in housing starts will only serve to prolong the housing recession, not help to curb it. Hopefully the pick up in May is a one month phenomenon.