-
Monthly price
~1k
Paid subscribers
-
Monthly revenue
0.0×
Paid trend
0.0×
Free trend

Popularity trend

Loading...
Get a weekly roundup of the best Substack posts, by hacker news affinity:

Top posts of the month

By hacker news affinity
day week month year all
kevinerdmann 0 likes 28 Sep 22
August new residential home sales are out, and to my eye, they continue to defy the reports of a collapsing housing market. Here are some charts updated from last month. First, homes sold, for sale, and months of inventory. There was a blip up in sales, which caused “months’ of inventory” to retreat sharply from last month’s high. There is little reason to look to months’ of inventory for any insights about the housing market, so it doesn’t really matter. Here, as always, it mostly just reflects changes in sales.
kevinerdmann 0 likes 17 Sep 22
A levelling off of home prices has accelerated the number of times I have seen accounts of sharply falling home prices in the media. There is a bit of a measurement problem with these reports. At any given time, you could find some incomplete estimate of home prices that could give you a range of +/- 5% or 10%, but aggregate prices just don’t change like that. Even in 2007-2010, prices at the steepest points of decline nationally were falling by less than 2% monthly, so it really is, even at its worst, like watching a slow motion train wreck. I suppose it’s possible that there could be unprecedented price trends in this cycle, but more likely when you see a report of prices in some segment or city declining suddenly by 5% or 10%, it’s a combination of compositional changes in the measurement, noise in the price data, the sentiment of the messenger, etc. There is a tradeoff between timeliness and relevance that is always difficult to balance, but I don’t think much is gained by putting credence in these volatile reports. The slower moving data shows the same trend shifts. It just doesn’t exaggerate them. So I don’t really see any loss of information by sticking with aggregate, slow moving numbers.