More on Portland Housing Supply …
Posted on March 28, 2008
Filed Under Portland, Statistics, Buying Real Estate, Real Estate, General |
Terradatum has just added a new search algorithm that gives a condensed and more up to date view of a given market: Six months, broken into weekly and twelve week increments. It came out yesterday so I haven’t had time to go over it in depth, but one chart stands out. This chart (click for PDF) verifies that inventory is beginning to shrink:
This shows the total properties for sale for even one day in any given week; the following week adds new listings and subtracts expired listings and listings gone under contract…
UPDATE: Finally getting an hour to look closely at the numbers, I’m going to side (for the time being) with the skeptics, and apologies for posting before I was convinced. Nothing is more important than accuracy, and while the trends comport - inventory reached its peak last September, residual inventory has been dropping since; new listings are down YoY in March and expireds are up - the numbers don’t make sense: There are currently 15,200+ active listings. I’ve had a call into Terradatum since Friday; will follow up when I hear back.
UPDATED UPDATE: Whatever Terradatum did in their upgrade, in cross referencing RMLS numbers it’s thrown most of the data points off by up to 30%. The month end stats - which were posted yesterday - are off as badly as the six month stats. I’ve talked to customer support and sent documentation, but typically CS is a filter for the techs, who tend to think automatically that we just don’t understand what we’re seeing.
I hope they correct it; it’s been a terrific tool.
Comments
12 Responses to “More on Portland Housing Supply …”
Leave a Reply
This is an odd chart indeed - it goes against everything I would expect in a spring market and I’ve seen from other data sources. We seem to have ~=900 properties a week coming on the market and about ~=300 sold yet inventory has dropped by 30% ?
Strange indeed - is there an available link to Terradatum or is it a pay to play service?
I’d like to take a closer peek at their methodology as their numbers seem odd on the face of it.
This data looks fishy for a number of reasons.
1. The scale of the chart is compressed to make a small change look huge.
2. More importantly, as Uncle_Git points out you can’t have shrinking inventory when you have 2x as many new houses coming on the market as you do sales. Even including listings going under contract it still doesn’t add up.
3. This RMLS data ( http://www.housingtracker.net/askingprices/Oregon/Portland-Vancouver-Beaverton/ ) as well as the RMLS data I’ve seen clearly shows a continued increase in inventory.
I’d do some digging into this data Jeff before you distribute too widely.
Can’t you two see that the data clearly screams “BUY NOW!” ?
“We seem to have ~=900 properties a week coming on the market and about ~=300 sold yet inventory has dropped by 30% ?”
Not sure if there’s any accuracy in this data at all, but perhaps one explanation is that people are pulling homes off the market that have been on for a while and they’re renting them?
Seems to be some evidence that it’s happening - I’ve seen this on my daily commute (of course that’s anecdotal); some houses that have had forsale signs in front of them since late last summer now have for rent signs in front of them.
Sellers either giving up and trying to get some kind of cash flow, or perhaps thinking that they can get a better price next year. Pent up supply?
Could be - but 600 a week deciding to become enforced landlords ?
Seems high to me - you could I suppose start doing a weekly search on craigslist and see how the “for rent” inventory is going.
If 600 people a week are deciding to become landlords I’d expect to see rent dropping - making the rent Vs Buy comparison that much more of a no brainer.
Please see update: I’m questioning the numbers as well.
To be fair - the fact that they are trying to make a living selling the data would lead you to believe that the data was at least accurate
See updated update…
Thanks for the update Jeff - I’d agree - it does look like a useful tools from the demo I took a poke at on their site.
Any guesses on how March stats will shake out regarding sales ?
I know it’s a bit early as sales will trickle in for a few days yet.
Git, March numbers are still being posted, but it looks like more of the same: unit sales down 30%-35%, median price up a percent (which will be the lede). Expirations are UP over 60%, and new listings down about 8%, so - and this is a guess - MSI, as calculated by RMLS, will be in the eight month range.
Not pretty; not dire.
Do you think people are letting their listings expire so they can get a new RMLS number and a reset of “Days on Market” for spring season or do you think the owners are really taking them off the market to rent them/decided not to sell ?
>Do you think people are letting their listings expire so they can get a new RMLS number
No; that happened at the end of last year and the beginning of this. If it were the case, you’d see an uptick in new listings. Anyone who doesn’t have to sell is pulling out.
Incidentally, just got off the phone with Terradatum; all fixed, will post shortly.