Oct 6 2006 12:18PM | Permalink | Email this | Comments (0) |
Blog This! using: Blogger.com | LiveJournal |

The last few — and the next few — monthly economic and construction market reports have to be interpreted cautiously. During a transition to slower growth initial estimates get revised more substantially. The nonresidential building market is excluded from this caution because the strong growth trend has not changed.

Early in the transition period the initial estimates are typically revised lower. That has clearly happening in the last few months for housing and civil construction activity and also for consumer spending. This occurs when missing data is estimated too high because the estimating routines rely on trends over a period longer than a few months. Then, as soon as the statistical agencies correct this source of error they begin to estimate too low because the last 4-5 months are now a period of below trend market activity. Expect data revisions to switch from generally down during the summer to generally up during the fall.

Buy the middle of next year we will learn that the housing and economic slowdowns in spring and summer 2006 were deeper than reported at the time and that the fall period following those slowdowns was stronger than portrayed in initial market data reports.

But it not just measurement problems during an economic transition period that makes monthly market reports misleading. Even accurate measurement practices would yield an underestimate of sustainable market activity levels during a period of transition to slower growth because some demand is met from inventory. The latest housing starts report illustrates this. The 1.665 million starts in August were low relative to new home sales. Starts could dip below the August level briefly in the next few months without us changing our view that sustainable housing demand remains over 1.7 million.

It would be misleading to see the low August housing starts total, and possibly an even lower total in one of the following few months, as a sign that starts will decline progressively to a much lower level. Focus on changes in the surplus inventory of homes, even though this data is much less reliable than data for either starts or sales, so that you can estimate the demand for homes.