Conversation
![David Watson 🥑](data:image/svg+xml,<svg xmlns="http://www.w3.org/2000/svg" width="96" height="96"><rect fill-opacity="0"/></svg>)
Post your reply
The American Housing Survey also doesn't show a major run-up in the median age of average buyers *or* the median age of first-time homebuyers.
This is average, not median age, but the New York Fed finds that first-time homebuyers were *younger* in 2024 than in the 2000s. Haven't gotten older on average in nearly 20 years.
h/t
In Agency MBS data, the majority of first-time homebuyers are age 34 and under.
That means the median first-time homebuyer is 34 or younger. Hasn't changed a lick post-COVID.
As my colleague has pointed out, Gen Z and Millennials have taken longer to achieve the same home ownership rates than prior generations.
Homeownership is indeed less common for young people than it used to be.
Just isn’t nearly as dire as NAR implies.
Quote
Ben Glasner
@BenGlasner
Replying to @BenGlasner
When you track actual homeownership by generation, defined clearly and consistently, you see something else.
Millennials and Gen Z are behind Boomers, but they are converging toward Gen X. Also, that gap is not nearly as big.
New York Fed Consumer Credit Panel data from Equifax (which excludes all-cash buyers, TBF) finds that a majority of home buyers were first-time buyers in 2023.
Overall volumes are down since the early 2000s, of course.
Sorry, this is my obsession this week. Can't do anything about it now. The train has left the station on this.
Well, the gold standard surveys and agency MBS data are all singing the same tune. NAR is a huge outlier, but only post-covid. My guess is something about their survey has changed or degraded.
Would make sense, if they're buying second / third homes, or more easily trading one for another. The first time median hit 40 didn't it?
The NAR says the median buyer is 59 and the median first-time buyer is 40. Highly skeptical of these numbers Need to test the latter using the AHS.
so... who has the more reliable data? the realtors (with MLS), or the survey?
Always Be Checking Your Data Generating Process
Quote
ascenditque
@ignisdepetra
Replying to @besttrousers
Just went back and looked, and I think this is a measurement problem, because that is too extreme to be organic. Turns out HUD and Fannie/Freddie define "first time homebuyer" as not having ownership in a primary residence for 3 years prior to purchase.
Is it all relative to the prior survey data though? So it's spiked up a lot compared to their earlier findings.
Either way, I expect this spike to be very short-lived and return to more normal levels.
MBA should have the best info since they collect age data during the process.
The ACS sampling frame is the resident of the household/address. It doesn't provide the age of the landlord who owns each single-family rental.
Which one is more accurate?
What could explain the difference? Sample size or representativeness? Accuracy of age data?
"Moving" doesn't include the 5 rental houses that each boomer bought and sold last year. That's why.
It’s true. The economists and tech folks got data drunk and missed their entire hypothesis.
Even manipulating regulations. How dumb can people be?
Dumb enough to crush an entire buyer sector built to save the day.
Don’t worry the 50 year mortgages will drop that number and help younger generations pay 20 years of interest 
There's a lot to unpack there. BUT, as we had a housing boom a starting a few decades ago, and people are living longer, there are a lot of retirees who sell their homes and buy new ones to live in for retirement. This would drive up the median.
