The assumption for a couple of generations has been that housing costs will always increase over time as a percent of wages. The problem with that is that it's unsustainable. The next generation has to be able to afford it in order to buy it, and they also need to buy food and utilities etc., which also cost more when real estate does. Investors can't save you either when the next generation can't afford the rent they would have to charge to turn a profit.

But in order for housing to be a good investment, it has to have competitive returns with other investments, i.e. it needs to increase by at least as much as GDP per capita. Meanwhile median wages have been increasing slower than GDP per capita, which as above is the long-term cap on housing prices. In other words, housing can't long-term sustainably beat other market investments unless wages do, which they haven't, in which case people would get better returns by putting their money in stocks etc.

Worse, years of ZIRP inflated housing prices beyond any sustainable level even with the scarcity being maintained by existing zoning restrictions, i.e. the "eventually it's not sustainable" point is already in the past.

The result is that in order for housing to be a good investment going forward from now, there would first have to be a major housing crash so that "investors" (i.e. home buyers) could buy low instead of buying high. Which thereby implies that it wouldn't be a good long-term investment at current prices. And by major housing crash, notice what "enormous housing bubble that crashed the world economy" looks like on this chart in 2007 and compare it to what things look like since then, especially since 2020:

https://fred.stlouisfed.org/series/MSPUS

A lot of people haven't realized that the party is over and are expecting housing to still be a good investment.

That's the bank's problem.