Nov 20 Market Update

Weekly sales numbers courtesy of the VREB.

Nov 2017
Nov
 2016
Wk 1 Wk 2 Wk 3 Wk 4
Unconditional Sales 83  241  408  599
New Listings 157  387  553  786
Active Listings 1909  1916 1855 1815
Sales to New Listings  53%  62%  74%  76%
Sales Projection  569  616
Months of Inventory 3.0

The case of the misleading sales statistics continues from last week.   Not quite as bad this week, but still of the 167 sales reported for the week, 8 actually happened in the spring, and another 11 happened between the summer and 2 weeks ago.   So while sales appear to be up 11% from last year, they are actually going at about the same pace.   Still a quite active market though.   We will likely pass 600 sales which is significantly higher than the 10 year average of 484.   In a slow year we would see only about 380 sales in November.

Continue reading

Nov 14 Market Update

Weekly sales numbers courtesy of the VREB.

Nov 2017
Nov
 2016
Wk 1 Wk 2 Wk 3 Wk 4
Unconditional Sales 83  241  599
New Listings 157  387  786
Active Listings 1909  1916 1815
Sales to New Listings  53%  62%  76%
Sales Projection  569
Months of Inventory 3.0

Quite a strong week of sales to put us at the same sales pace as this time last year.   Or is it?  I was surprised at the high sales this week, which more than made up for the weakness in the first week and put us above the sales rate of this time last year which we haven’t seen for quite some time.   However my job is not to mindlessly post stats, so while searching for something intelligent to say about the market, I picked through the week’s sales.

As a reminder, the weekly numbers released by the VREB are not the sales from the week, but rather whatever listings were marked as sold that week (the complete and completely boring explanation is here).  Usually those sales are from the week or the previous one, but sometimes the sales that are entered are really old.   Like this week, where included in the tally were some sales from not last week, not the week before, not the month before, or the one before that, but sales from last year.  And it wasn’t just a few, it was 20 condo units of a new build out in Langford that sold a full 11 months ago but showed up in the weekly numbers just now.

So no, sales are not up compared to last year, they are down.  This just cements my belief that we need to get off the silly reporting date based numbers and get on an accurate count of actual sales pronto.   Luckily that is almost ready.

Continue reading

Dark sales

All the data on this site are based on the sales and listings in the MLS database.   There are just under 6000 of those sales in a slow year, and over 10,000 in an insane one.

Thing is, while most sales happen on MLS, there are quite a few that are not captured there.  For new builds, only a small fraction are ever listed on MLS, and none of the private sales would show up there either.

How many sales are happening outside the MLS system?  Well one side effect of the province publishing foreign buyer data is that they are also publishing the total number of land title transfers every month.   Now that we have a full year’s worth of that data we can get a sense of the number of these dark sales.  Because sales are recorded when the contract goes unconditional, and title transfers happen on closing which is usually a couple months later, I’ve advanced the title transfer data to best fit the seasonal pattern. Continue reading

Nov 6 Market Update

Weekly sales numbers courtesy of the VREB.

Nov 2017
Nov
 2016
Wk 1 Wk 2 Wk 3 Wk 4
Unconditional Sales 83  599
New Listings 157  786
Active Listings 1909 1815
Sales to New Listings  53%  76%
Sales Projection
Months of Inventory 3.0

20% below last year’s daily sales rate but that doesn’t mean much since we only had the tail end of last week in this week’s numbers.  Will require another week or two to determine whether the stress test is pulling buyers forward, or whether the increasing number of lenders that are already applying it will start to put a damper on sales.

Continue reading

A foreign buyer update

A couple days ago we got some stories about how the foreign buyer numbers were finally released to cover the period from June to September 2017.  It was a bit strange, since every story referenced “statistics released Tuesday afternoon by the province” but I couldn’t for the life of me find those statistics.  There was no press release by the Ministry of Finance like we had before, and the data was not up to date on DataBC.   So was the data released only to journalists?   Can anyone help me find this so called release?

No matter, after some gentle poking at the ministry and the help of a friendly HHV reader, the DataBC record was updated with the up to date foreign buyer data this morning.

Almost all the data reported in the media about Victoria is slightly wrong, because they divided the number of foreign buyer involved transactions for all types of properties (residential and commercial) with the total number of residential transactions to get their percentages.  No big problem, they were only off by a tenth of a percent or so.  Here is the up to date data.

No great change in the numbers for our region.   I still find the numbers for Saanich surprisingly high at 8.3% for the year so far, nearly twice the level of Victoria proper.   It doesn’t combine well with Saanich council dragging their feet on every new development.

October Update

Every month I try to get a jump on the VREB by pulling the data for sales and new listings and prices from the VREB database before the official numbers are released.  And every month I’m frustrated by the fact that it is impossible to match the numbers that VREB produces.

The problem of course is all of your insatiable thirsts for immediate information.   Monthly stats are compiled the morning of the 1st of the following month, and the problem is that quite a lot of sales for the month have not been entered into the system yet.   So what do they report for the month’s sales?  Well all the sales that were reported during the month, minus the sales that were reported in previous months but collapsed.   Problem is that means what gets reported as sales in a certain month are actually missing quite a few sales at the end of the month, and include several sales from the previous month.   For example, as I write this, of the 280 single family sales that the database says were reported in October, 15 actually happened in September, and the total reported by the VREB is 294.

You’d think counting sales wouldn’t be so difficult.

Maybe it doesn’t matter, but fundamentally data that depends on how prompt agents are at entering their sales isn’t very reliable.   Much better to just throw the whole reporting dates out the window and move to a model where everything goes by the date a sale actually happened.   Collapsed sales are an insignificant number and can be tracked separately.   As of now I can actually do that since I gained access to the raw data feed from the VREB and am using it to build out some better reporting tools for the market.  Along with that I will at some point transition to statistics based on sales dates rather than reporting dates and thus will diverge from the published VREB stats somewhat.   Stay tuned for that.

For now, we’ve got the stats as they come from the VREB.  And this month is much the same as September.   Overall sales down some 10% from last October with detached declining the most, and condos maintaining the feverish pace from last year.


Continue reading

October 30 Market Update

Weekly sales numbers courtesy of the VREB.

Oct 2017
Oct
 2016
Wk 1 Wk 2 Wk 3 Wk 4
Unconditional Sales 146  289  453  601  735
New Listings 265  479  698  922  904
Active Listings 1990 1989 1978 1982  1938
Sales to New Listings  55%  60%  65%  65%  81%
Sales Projection  565  661  624
Months of Inventory  2.6

Running about 15% behind last year’s sales rate.   Today we saw another 36 sales entered (some of those being from up to two weeks ago) and we’ll see likely another 30 tomorrow for about 660 this month.

Continue reading

No, we aren’t underbuilding in Victoria and other census bits

Housing stats are out from the census, and with it a bunch of news articles about what can be gleaned from the release (aren’t we all glad the long form census is back?).  In Victoria the ownership rate is just short of 63% which is down a couple percent from 10 years ago.  The breakdown by housing types is as follows.

Why is the duplex category so big?  It’s because every basement suite counts as a “flat in a duplex”.

Also some other stats I thought were interesting, on the age of our housing stock:

Too bad they didn’t break down the years before 1960 and the bins are different sizes, but you can clearly see the large building boom of rental apartments along with neighbourhoods like Gordon Head in the 60s and 70s.

57.1% of owners have a mortgage, which is a smidge higher than the 55.7% it was 10 years ago.   The percentage of owner households spending more than 30% of their income on housing is actually down to 19% from 20.9% 10 years ago which indicates fewer households were stretched financially due to their housing costs.  Probably something to do with the fact that in 2016 mortgage rates were half those of 2006.  The census was also before most of the recent runup in housing prices.

Back to construction for a minute though.  The usual suspects are trying to present the data to say that construction has not kept up with population growth (clearly we need to build more!).  In the last 5 years this is true.  We had a growth of 23,455 people and only added 5330 housing units, which would yield 4.4 people per unit, which is well over the average household size of 2.2.   But housing construction goes in cycles and 2011 to 2016 was a slow period for construction while population grew faster than normal.   That’s quite obvious looking at construction history.

Looking back 10 years, we grew by 37,682 people and added 17,335 dwelling units, which yields a household size of 2.17 people, or pretty much exactly our average.   No evidence of systematic underbuilding.

One look around the city (or a glance at the stats) would also show you that construction has picked up massively from the levels of 2011-2016 so I don’t think we need to concern ourselves too much with running out of housing quite yet.

Oct 23 Market Update

Weekly sales numbers courtesy of the VREB.

Oct 2017
Oct
 2016
Wk 1 Wk 2 Wk 3 Wk 4
Unconditional Sales 146  289  453  735
New Listings 265  479  698  904
Active Listings 1990 1989 1978  1938
Sales to New Listings  55%  60%  65%  81%
Sales Projection  565  661
Months of Inventory  2.6

A surprisingly strong week at 164 sales which is 20 more than the first two weeks of the month.  I wonder if the stress test is pushing some people to jump in before they can’t qualify anymore?   A 25% buying power reduction is nothing to sneeze at.  We started out the month at a sales rate 30% lower than this time last year, now we are just off 10%.

Continue reading

Oct 16 Market Update

Weekly sales numbers courtesy of the VREB.

Oct 2017
Oct
 2016
Wk 1 Wk 2 Wk 3 Wk 4
Unconditional Sales 146  289  735
New Listings 265  479  904
Active Listings 1990 1989  1938
Sales to New Listings  55%  60%  81%
Sales Projection  565
Months of Inventory  2.6

143 sales for the week which is about the same as the 146 in the first week.   However the year over year drop has moderated significantly, now down 23% compared to this time last year.   That sales rate puts us on track for about 565 sales for the month which would be exactly the 10 year average of 567 for October.

Continue reading