TV-Viewing U.S. households have increased at the slowest rate in 10 years, says Nielsen Media Research. The number of television households is 114.9 million, Nielsen says. The immediate conclusion some observers will leap to is that Web video is displacing TV watching.

 

But there are simpler explanations. The number of occupied housing units has a direct bearing on the number of homes using TV. The number of people in a country affects demand for housing. Population growth, new housing starts and general economic conditions all play a role in setting the natural rate of growth for TV households.

 

One simply has to note that the rate of new home construction has fallen sharply between 2006 and 2008. That means fewer new homes, fewer new occupants and fewer occupants watching TV. Where in 2007 there were about two million "new housing starts," in 2008 the rate had dropped to one million and in early 2009 seems still headed further lower.

 

But one also has to factor in changes in vacancy rates. Fewer existing dwellings occupied, the fewer will contain people watching TV. Also, some homes are not occupied on a full-time basis. Vacation homes or second homes are an example of that. Then there is housing removal, as some older units are scrapped.

 

It appears that the household formation, a driver of housing demand, slackened between 2006 and 2009, by perhaps 400,000 units a year in late 2008.

 

Also, during recessions in the mid-1970s, early 1980s and early 1990s, housing formation also slowed, so there likely is a recession effect here.

 

Also, the Congressional Budget Office notes that the percentage of vacant units has risen since 2006, and is directly attributable to foreclosure and sluggish retail activity, which is keeping some units from the market.

 

About 37 percent of vacant housing units are seasonally or occasionally used. Vacation and second homes largely drive this category. At any given time, about six percent of housing units are rented or sold but are yet to be occupied.

 

About 17 percent are vacant for some other reason. In the second quarter of 2008 about 12 percent of vacant units were for sale while 21 percent were for rent.

 

Between 2008 and 2012, increased household formation will largely be driven by changes in population, the CBO says.

 

A TV household is defined as a home that has television sets turned on at some point during a measurement period. Even if users are watching some video on mobiles or PCs, it is unlikely a significant or even measurable number have ditched their TVs, do not use them at all, or have terminated a multi-channel TV subscription.

 

In other words, most of the mobile and PC viewing is supplemental, rather than a complete functional replacement.

 

There does not seem any credible evidence that such behavior--scrapping TVs or video subscriptions--has happened at anything more than a frictional rate so far, at any rate.

 

Under those circumstances, it seems wisest to assume the changes in TV household growth are related more to the declines in housing starts, household formation and increases in vacancy rates than any measurable changes in TV viewing habits.

 

Of course, it would make for a better, catchier story if we could point to some significant change in TV habits caused by Web substitution. It just doesn’t appear that the Nielsen data is best explained that way, though.