Americans may be watching more TV than ever, but all that viewing is happening in fewer homes.

The 2011-12 season marks the first on record with Nielsen to see a year-over-year decline in the percentage of U.S. households with at least one TV, known as the TV penetration rate. For the first time in 20 years, the total number of homes with TVs — what Nielsen refers to as the universe estimate — has dropped, from 115.9 million last season to 114.7 million.

With the overall number of U.S. households (with or without TVs) rising to 118.6 million over the same period, the TV penetration rate dropped 2.2% to 96.7% — the lowest since it hit 97% in 1975, according to Nielsen.

“What’s staggering to me is for decades the TV penetration rate has been at 98%-99%, so this is a pretty notable dropoff,” said Brad Adgate, senior VP research at Horizon Media.

Popular on Variety

The percentage drop means nearly 5 million viewers have vanished, leaving a universe of 289.7 million.

So where did 1.7% of the viewing population go?

The reasons for the contraction of the TV world are not entirely clear. The first theory: In a tough economy, Americans don’t buy as many TVs, a trend compounded by the 2009 digital transition that forced consumers to upgrade or replace their equipment.

Because the old analog sets are no longer counted, Nielsen immediately saw a downturn in TV homes once the transition deadline transpired, and though the numbers eventually rebounded, they still haven’t returned to pre-2009 levels.

“The drop in TV households after the digital transition was undoubtedly tied to a poor economy,” said Pat McDonough, senior VP of insights and analysis at Nielsen. “Most homes that dropped TV were younger, poorer homes.”

She also noted that Nielsen has periodically revisited non-TV homes and found that they have double the vacancy rate of TV homes — indicating that people in non-TV homes move frequently, possibly because of worsening financial conditions in the household.

Blame the recession

The recession may have also stunted the growth of TV households because young adults who would have gotten jobs and residences of their own are increasingly moving back in with parents after college amid high unemployment rates.

“Kids who were normally in an apartment in (New York City) and counting as a new household were now staying home with ma and pa,” said David Poltrack, chief research officer at CBS Corp.

But it’s also possible that what may seem a cyclical economic trend is also a secular, technological one. The decline could be tied to the highly disputed phenomenon known as cord-cutting, in which a segment of the population is bypassing multichannel subscriptions to watch on digital platforms.

Just how many consumers may be cord-cutting is subject to debate. Many studies have shown their ranks are statistically insignificant, but many analyst ascribe the sizable dips in subscribers that cable operators saw in the second quarter to cord-cutting.

Just a phase?

Nielsen has found that online viewers tend to be recent college graduates in big cities. While poor, rural markets over-index among decreasing TV homes, there are tech-savvy, college-heavy urban centers that also show marked decreases like top-10 market Boston, down 3.3%.

The big question is whether those watching online are doing so because they can’t afford new TVs or because those TVs have become an anachronism to them.

“We’re curious to see if this is simply a lifestyle phase that they’ll grow out of as they get the house and the kids,” McDonough said.

Nielsen research in May and July showed an unusual increase in broadcast-only homes, suggesting that rather than simply tossing aside TV entirely in favor of laptops or tablets, viewers may be attempting a combination of free over-the-air broadcast TV and such paid digital services as Netflix. That trend is playing out at the same time that subs disappearing from cable are not flocking to satellite or telco options as they’ve done historically.

Nielsen doesn’t think cord-cutting is entirely the answer, but that hasn’t stopped the company from exploring how to extend its measurement of digital platforms in TV homes to non-TV homes, which would then be factored into universe estimates.

Nielsen has made inroads into tracking usage on digital platforms, but integrating those stats into TV homes numbers is a tall order and won’t be completed this year.

“This is a complex business discussion we need to engage in with our clients,” McDonough said. “We have started this conversation and will continue it as we evolve our measurement across platforms.”

Someone has to pay

Whether the drop is an aberration won’t be known for some time. But if it is the beginning of a long-term trend, there will be consequences for what advertisers pay to reach audiences. For now, though, the new numbers were factored into negotiations given that Nielsen notified the industry back in May.

“In the upfront, everybody just built in the numbers, and it was business as usual,” Poltrack said.

The declines are playing out across all demographics and also clearly suggest an aging of the audience that shouldn’t play well with advertisers who target the under-50 crowd. The 18-49 demo slipped from 131.5 million last season to 127.9 million in 2011-12.

Nielsen was expecting some sort of change in the 2012 universe estimates given that it would be the first to incorporate population measurements from the 2010 census. Every 10 years the universe estimates are recalibrated to take into account the latest census figures. But in 2002, the census actually spurred a 2002 increase; 1992 was the last time universe estimates went down, but they shot right back up over the remainder of the decade, which McDonough credits to the country shaking off a recession quicker.

“The boom in the 1990s helped TV households rebound at that time,” she said. “We will certainly be watching to see if history repeats itself.”