Comment by pessimizer | Hacker News Reader

Comment by pessimizer | Hacker News Reader

Parent Full thread

"The S&P CoreLogic Case-Shiller 20-city home price index, released Tuesday, soared 17% in May from a year earlier on top of a 15% jump in April. The May increase was the biggest since August 2004."

This article is garbled, as far as I can tell. What I think they're saying is that from May 2020 to May 2021, prices rose by 17%. So that wouldn't be on top of the 15% from April 2020 to April 2021, it would be overlapping with most of it. It also wouldn't be a "May increase" but a yearly increase.

They're talking about it as if it were a monthly price rise (or an annualized monthly price rise to be generous to the author), but if it were, what does "a year earlier" have to do with it?

The worst possible interpretation is that the headline is saying that prices rose 17% faster this May as compared to last May (e.g. prices rose 0.6% last May but 0.7% this May) but that wouldn't be a story.

edit: also, wasn't May 2020 around peak covid terror time? I can't imagine people were looking to make long term investments.

This confusion happens over and over in popular press reporting. To the point where unless it's explicit I now assume it's annualized.

I assume it is intentional to get more clicks.

Sales of anything tend to be tied directly to human action which is more or less seasonal. So it makes sense to compare to a comparable time the year before.

I like millions of parents spent a few hundred dollars on backpacks, shoes, pencils and lunchboxes this week for back to school. Won't be doing that again until next year.

The closest reset (when stores update their prices/products they carry) to November this very crappy wine I sometimes buy, goes from $6 a bottle to $3.89-- presumably to captures sales for holiday get togethers where people are looking for quantity and not quality. It stays at that 3.89 until after New Years.

I don't know what the cycle is for housing-- but I suspect under normal circumstances sales would be highest about an escrow-long time before summer vacation so moving would be least impactful on kids.

It’s definitely annualized. That makes the most sense. And when you look at the case shiller index directly it bears that out. Also interesting is Feb to Feb was only 12% annual increase. But may to may was 17% increase so there was definitely an acceleration there. Some markets are insanely hot. We saw homes going for 30-40% more than the previous year in our neighborhood. A house almost exactly the same as ours sold this year in June for almost double what we paid for ours a few years back. Some real estate markets are more than just hot.

e: its rolling 12, which is what I meant, but yeah.. words matter :)

I looked it up and it's not annualized, it's year over year. Therefore very sensitive to chosen start and endpoints.

OK, but with year over year, it makes exactly zero sense to say that the May-to-May change of 17% was "on top of" the April-to-April change of 15%.

Maybe they meant it’s “on top of” it in a table of YoY changes by month sorted in descending order ¯\_(ツ)_/¯

wasn't May 2020 around peak covid terror time?

Yes, but that's when people started to want places outside of big cities.

which would depress prices

> wasn't May 2020 around peak covid terror time?

I bought my current house in May 2020. We got a ton for our old house due to people starting to move to the burbs, but still paid a good price (relative to today) for our current house. We thought we did pretty well in the grand scheme of things.

This is the one case I would believe if it was month over month growth. Rents were so insanely low in Jan-Apr (some places even 50% off). Now everyone is trying to get a place.

Reply on news.ycombinator.com