Housing construction limping along

People were chattering on Friday about the "shocking" housing data. New residential "starts" were down -14.3%. Yeah, that's a really bad number, but is it always nest not to take one data point in a series and take it out of context and draw strong conclusions from it. Here's a little more context from the official New Residential Construction report:

1. Permits declined only -2.8% in January compared to December and were still well above the level of October and November.
2. Housing units that had previously been given permits but not yet started were actually up by +2.9%.
3. Starts declined by a whopping -14.3% (or an even more eye-popping -37.8% from a year ago), but were only -4.7% below the pace in October.
4. Housing units under construction declined by -2.4%.

You need to look at the full picture, with all four data items, and you need to track them for more than a month or two. We all have the separate data on new housing sales and existing home sales, as well as pricing for new and existing homes. We also get weekly data on mortage applications, both to purchase a home and to refrinance an existing mortgage. You need to look at all of this data to get a more accurate picture of the economic impact of the housing market.

My expectations from a few months ago remain unchanged: the housing pullback is likely to take a few more months to wind down, somewhere between February and May, before beginning a slow rebound.

Wall Street and so many of today's commentators and pundits and journalists are extremely prone to knee-jerk reactions. Be careful to make your own assessments rather than get sucked up in their hyperbole (positive or negative).

-- Jack Krupansky