The Yesterday we saw the release of the July New Construction data. Some sources reported that the data was pretty good, with headlines such as "New construction climbed to a six month high in July" and "Housing Starts Climb to Five Month High."  This column had a slightly different take. Yesterday's column was "July New Home Construction Recovering Still." It revealed that the July New Housing Starts and Completions were Slower than 1981 through 2007. Stated another way, New Home Construction has not returned to pre-recession levels.

The new home market is still in recovery mode. There was also some data being shown that we were on track for over 1 million starts for the year. This did not make sense. How are we doing with regard to starts and completions, on a rolling or running year basis? How are we doing with regard to Total Housing Units Started and Single Family Units Started?

Both Completions and Starts are turning downward. This is not being reported elsewhere. If you compile the data and produce a running year or rolling year  number you find that even though our completions and starts were higher than last July (July 2015) we are seeing a decline in the unadjusted number.

We are seeing Fewer Annualized Starts and Completion than we Saw during 1983 or 1992.  We saw years of growth after July of 1983 and July 1992. The total units started have seen a drop in the annualized rate of growth. The single family starts are continuing to rise. This means than non-single family residential units (Condominiums and multi-family properties) are declining.  We have seen a surge in non-single family homes after the housing recession. Fewer people are buying homes. A lower percentage of people are home owners. A higher percentage of people are choosing to rent instead of own their residences.

Single Family Starts are Trending Slower Than 1983 and 1992.  We saw over 1 million starts during 1983 and 1992. We saw roughly 400,000 units started during the final 5 months of 1982 and 1993. If we add 400,000 units to the July figure that places us at 860,000 units. If the starts are slowing then we may barely exceed 800,000 units.

The Total Housing Units being Completed are Slower than 1983 and 1992, as well.  We saw between 1.1 and 1.4 million completions during 1983 and 1992. The completions level for 2016 is significantly lower than either of those years. We saw 500,000 and 700,000 units completed during 1983 and 1992, respectively. If we add 500,000 units we will barely eclipse 1 million units completed. If this segment is slowing, as is believed, then we will fall short of 1 million units completed.

The housing market is part of what our US economy is based. Retail sales pick up when houses are built and sold. More house means more construction jobs. A weak new home market means a weak economy.

Where are we heading? Weak new contruction data should continue. This means weak new construction sales numbers. This means we may be seeing the beginning of a weaker economy and possibly a recession.