Housing starts for November were released today, December 20th, and the stock market rallied strongly on the supposedly "good" news. The statistical error rate in the housing report is so huge, that the numbers are meaningless -- and easily subject to manipulation by a government that is desperate to provide news of a recovering economy.

Housing starts peaked at 2,273,000 in January 2006. According to the Commerce Department, construction of new U.S. residences in November 2011 was 635,000. Almost five years later, housing activity is still less than 28% of what it was at the peak. Despite this almost three-quarters decline in housing activity, this is being spun as evidence of an economic recovery by the mainstream media. Would you consider it progress if your salary was only 28% of what it was five years ago?

As dismal as this statistic is, it is very possible the actual number is much worse. The housing starts report has the highest statistical margin of error of any government report. The error is so huge that is a waste of taxpayer money to produce this report. The error in the overall number can be greater than ten percent. The error on individual components can be as much as 33%. This is important because better housing start numbers in 2011 (November was not the first month when better numbers were reported, this took place earlier in the year as well), have been created by a supposed surge in apartment house construction. Apartment construction rose by 25.3% in November and this is what is making the overall number higher. Considering the huge statistical error rate, it is possible that it didn't rise at all.

Optimists who think it did rise by that much need to ponder the implications of why a lot of apartments are being built and very few single-family houses. The inescapable conclusion is that few Americans can afford to buy their own home anymore. I would hardly describe that as an indication of better economic conditions. Even more telling is that the number of completed housing units, which dropped by 5.6% in November even though housing starts rose earlier this year. If this is correct, a lot of housing that is begun is not being finished. While that makes no sense, nothing else about the report does either.

Disclosure: None

Daryl Montgomery
Author: "Inflation Investing - A Guide for the 2010s"
Organizer, New York Investing meetup
http://investing.meetup.com/21

This posting is editorial opinion. There is no intention to endorse the purchase or sale of any security.