12

Real Estate Reality: Despite the Implications, We Are Building More of It

The Seasonally Adjusted Building Permit and Housing Start Numbers Are Painting a Misleading Picture

James Brumley
The media spoke, and the people listened, but did the media explain it right? Perhaps, but as time passes and industry jargon becomes mainstream - even if not fully understood - sometimes the bigger picture can get lost in the shuffle. Wednesday's announcement of seasonally adjusted building permits and housing starts is one such possible example.

The monthly data, gathered and disseminated by the U.S. Census Bureau, have largely been stripped down to a couple of "bottom line" numbers... the seasonally adjusted annualized building permit requests, and the seasonally adjusted annualized number of housing starts. The former rolled in at 550,000 for October, while the latter totaled 519,000.

Both were considered disappointments, whether in comparison to expectations, to September's numbers, or even to October of 2009's seasonally-adjusted annual rates. Neither was a disappointment for the reasons widely assumed, though.

While the terms "seasonally adjusted" and "annualized" would imply the figures were some sort of average number intended to allow for a fair now-and-then comparison, in this case the mathematics may have misled investors by at least overstating the depth of the ongoing problems.

Take the actual, raw number of non-seasonally adjusted permit requests last month as an example. The total of 43,700 was indeed below the prior month's total, and it was even under the total 47,900 requested permits from October of last year. On a 12-month average basis, though (which works through the entire ebb and flow of the calendar and weather cycle), the total number of monthly permit requests has actually held steady near 50,000. That's in sharp contrast to the trailing 12-month average number of 49,200 permit requests seen in October of 2009.

The total number of actual housing starts paints an even more compelling picture than the one the media painted using the seasonally adjusted data.

How so? The adjusted figure of 519,000 for October of this year was 11.7 percent below September's adjusted number of 588,000. And it was even below last October's adjusted start rate of 529,000. Those numbers alone would imply that this October was an outright disaster, and an omen of a deteriorating real estate and building environment. The raw numbers tell a different story, though.

In actuality, the 44,300 starts seen in October of 2010 is right in line with last October's 44,500. In fact, the trailing 12-month average of housing starts currently sits at 49,100, versus only 46,700 last October.

Even more interesting is that we've truly seen more starts in the last twelve months (589,000) than in the 12-month period before this one (560,000). The same goes for building permits; nearly 598,000 permits have been issued since November of last year, while only 574,000 permits were issued between November of 2008 and October of 2009.

Said more simply, both of those numbers - which are real, raw figures - show progress, while the seasonally adjusted numbers indicated more weakness.

To be fair, all the permit and start numbers are currently in a downtrend. That's not an indication of weakness, though. That's just the typical slowdown around this time of year; it happens every year. The ultimate bottom should come in December. As long as none of the permit request levels or start numbers slide under 2009's lowest readings, though, we're actually on pace for a second year of measurable progress.

The rebound will be confirmed next year if we can exceed mid-2010's peak starts and permits. To be perfectly clear, though, the country has indeed built more real estate in the last 12 months than it did in the prior 12, no matter what the media may be suggesting.

None of this is to imply those adjusted calculations are invalid or unimportant; they do play an important role for decision-makers in certain situations. They are hardly the whole story, though, and unless the whole picture is grasped, we may come to the wrong conclusions about the construction industry's health.

U.S. Census Bureau, "New Residential Construction".

Published by James Brumley

Former stock-broker turned analyst turned writer, now helping investors who actually need good advice. I can tell you what the brokerage firms can't.  View profile

  • The U.S. has built more real estate in 2010 than it did in 2009, and is still going (fairly) strong.
  • Real estate activity tapered off between September and October because of seasonality trends.

To comment, please sign in to your Yahoo! account, or sign up for a new account.