New household formations finally seem to be picking up, but it’s not because young people are moving out of mom and dad’s house, according to a study released Monday by the Terner Center for Housing Innovation at UC Berkeley.

In fact, “age groups younger than 55 collectively had negative household formation between 2014 and 2015, while 65- to 74-year-olds accounted for more than two-thirds of overall household formation,” the report said.

A separate report released Monday by Zillow found that first-time home buyers today are older, renting longer, are less likely to be married and are spending a lot more of their income on homes today versus first-time buyers in the 1970s and 1980s.

Both reports add to the growing body of research on why Millennials are taking so long to strike out on their own, which is of great interest to Realtors, home builders and economists.

Historically, population growth alone results in the formation of about 1.2 million new households per year on average, said Jed Kolko, author of the Terner Center report. During the recession, this number “got down into only a few hundred thousand” as more people stayed with parents longer or found roommates.

Recent census data suggest that the rate of new household formation has started to improve, although there is conflicting data as to the start and size of the increase. The data cited most often showed a big year-over-year increase in the fourth quarter of last year, with gains continuing into this year.

It was assumed that Millennials, having found jobs, are now getting their own places.

But a recent report from the Pew Center showed that’s not true.

Even though unemployment among 18- to 34-year olds has dropped dramatically — to 7.7 percent early this year from 12.4 percent in 2010 — people this age are less likely to be living independently. Only 67 percent were living independently early this year, compared with 69 percent in 2010 and 71 percent in 2007, the Pew report said.

If young adults are not increasingly striking out on their own, then what accounts for the increase in new household formation?

Kolko attempted to answer this question by adding up the population growth between the first half of 2015 and the first half of 2014 for various age groups and multiplying by each group’s headship rate. Headship is the percentage of adults who head a household.

Like the Pew study, his analysis found that despite the job recovery, the percentage of young adults living with their parents continued to rise, even through June 2015.

He found that the headship rate fell or remained flat for all age groups from 20 to 64. “The only notable increase in headship rate was for 65- to 74-year-olds,” he said.

His conclusion: “Household formation is being driven by overall population growth and the aging of the population, but not a broad increase in headship rates. The recent gain in household formation appears to be due to basic demographic shifts rather than the economic cycle.”

What the data can’t explain is why young adults are living at home even after they get a job.

“One contributing factor is that 25 to 34-year-olds are decreasingly likely to be married or cohabitating with a partner, and — unsurprisingly — married or cohabitating young adults rarely live with their parents (just 2% do), compared with 31% of those who aren’t living with a spouse or partner,” Kolko wrote. The decline in marriage among young adults predates the recession.

Another possible factor is that “rent affordability is worsening broadly,” Kolko said. “When young people move out, it is usually to rent, not to own.” After the bubble burst, home prices fell precipitously, but rents did not. “Demand for renting rose as people lost homes to foreclosure or couldn’t get a mortgage,” he said.

Saving up for a down payment on a home is harder when rents are rapidly rising.

A third possible reason is that “the quality of jobs being created” is not good enough to let young people move out on their own, at least in the style they have become accustomed to.

First-time home buyers: Zillow’s report on first-time home buyers echoes some of the same themes.

It found that:

•First-time buyers are older than than they were in the past. The median age was about 32.5 in 2010-14, versus 29 in 1975-79.

•Americans are renting longer before buying a home. The median was 6 years between 2010 and 2014, versus 2.6 years in 1970-74.

•They are less likely to be married. Only 40 percent were married in 2010-14, versus 52 percent in 1985-90.

•They are spending a lot more of their income on a home. In the early 1970s, they bought homes that cost about 1.7 times their annual income. In the first half of this decade, it was almost 2.6 times.

Adjusted for inflation, the median income of first-time buyers has barely risen over this time period, from $52,831 to 54,340. But the median price paid for a first-time home, adjusted for inflation, has soared from $87,371 to $140,328.

It’s not clear whether home prices have simply risen faster than incomes or whether first-time buyers are buying more lavish homes, but it’s probably both.

“Some of what we are seeing is that Millennials are moving more toward the coasts, largely driven by employment. Homes on the coast are more expensive,” said Zillow Chief Economist Svenja Gudell.

Kathleen Pender is a San Francisco Chronicle columnist. Net Worth runs Tuesdays, Thursdays and Sundays. E-mail: kpender@sfchronicle.com Blog: http://blog.sfgate.com/pender Twitter: @kathpender

———

©2015 the San Francisco Chronicle

Visit the San Francisco Chronicle at sfgate.com

Distributed by Tribune Content Agency, LLC.