Housing starts are at ~1.7MM units/year right now, that's enough for about 4.5MM people at current household sizes. US population growth was 1.6M in 2019 and <1MM in 2020. Even with generous estimates on losses of existing housing, I don't think 3x is appreciably off.
Maybe it has something to do with the fact that more than half of the younger generation are currently living with parents [1]. Not because they necessarily enjoy it, but because they can't afford their own home.
Building more housing should lower the effective price of it, and allow many new home buyers or renters to enter the market.
A lot of the construction boom is probably attributable to the new found geographic freedom COVID gave, coupled with the low interest rates from the Fed's monetary policy. There was a sudden mismatch of locale supply and demand. Housing starts were also quite low over much of the last decade, definitely below the equilibration point for 2019-2013.
It'll be interesting to see how things play out. I think the shift of older millennials out of cities combined with the relatively smaller generation replacing them will likely cause a reversal of a lot of the urban price growth in the past decade in places like New York. Rents are already reflecting that, but sales prices have more hysteresis.
I've read since 2008 LA has added 5 jobs for every new unit of housing. I'm sure other places in CA must be just as bad with how they've been building for the past decade. It feels like we are so far in the hole in many cities, I'm not sure how much overbuilding it would take to right the ship and make housing affordable to the median wage earner again.