It is absolutely true that housing bubbles can span decades. But the longer they grow, the bigger the pop. Unless you deflate them first by increasing supply.
> In the last collapse, houses in "good" locations didn't lose any value.
The last collapse wasn't caused by the same things. What happened then is that banks loaned money to people who couldn't pay it back, which inflated housing prices until they started to default. The people who live in Palo Alto had better credit, didn't default, and could still get a loan even after the crisis (or didn't need one), so the houses in that area didn't lose value.
What's happening in this case is that housing in some areas is getting so expensive that even people making six figures can barely afford it. When it gets to the point that they actually can't afford it, the high end status of the neighborhood won't save them.
And a crash doesn't have to result in cheap housing. It can be a move from preposterously expensive housing to "only" very expensive housing. But if that's a 50% reduction in value, people are going to be unhappy.
> also I said they're not "universally" destructive, which I stand by, so long as some affordable housing options remain.
You may be right that if you consider only the most expensive houses, they may not lose as much value, because they're the ones that are actually worth ten million dollars. But non-universality is a small consolation if the effect still hits 85% of the local housing stock.
