Today I want to discuss the stats for May, and how they fit with previous months and years. It is clear that sales went up, but why? Let’s see what the numbers can tell us.
January to April recorded among the slowest sales in decades. As we discussed in our previous posts (1,2). For price changes, what matters is the Sales to Active Listing ratio (SAL), a measure of supply and demand. The following figure shows the SAL for Jan-Apr 2019 and how they compared to previous years. It is clear that it has been significantly below the 14% threshold we determined to be the threshold for prices to drop:
In contrast, May had a significant increase in sales, bring the SAL to 18%, the threshold for prices to not change much:
Although the SAL was 18%, that was because the inventory was lower than in the past. In terms of sales, May 2019 reached a 19-year low:
May – Price changes
It is clear that with a SAL of < 14%, prices would drop. Here’s the Composite HPI benchmark, which showed an 8.5% decrease Y/Y (nominal), or 10.9% if corrected by inflation:
If we correct by inflation, and plot per property type as percentage of the maximum value they got, here’s how it looks:
It is clear that the detached housing is the one most affected, which is consistent with that being most affected by the drop in sales.
Johnny Oshika has collected data that as far as we can tell covers most of the transactions reported by the REBGV since last March. So we analyzed it for trends in May.
(Disclaimer: We did not curate all the data and assume no responsibility for missing points or errors. We just want to get a read of trends. Take any specific information carefully and do your own analysis if you are in the process of making a big decision.)
The following plot shows the average and median prices with respect to the assessed prices. The trend is very clear. The Y/Y price difference is ~8%
Broken down per unit type:
Finally, the plot below shows a clear trend in which the percentage of units selling below the assessed price 2019 has grown to 74% in May. In other words, in May 2019, 3 out of 4 units sold below its assessed price.
Change in diligence time
The following workflow shows how a sale is processed and reported as we understand it:
A unit sale is reported by the REBGV by its entry date, which is within 5 days of the subject removal. It’s not reported by its sold date, the date when the offer was accepted (that’s why in the charts above I put a note saying that we still don’t have all the data for the sales from May).
The result of this is that many sales reported in May were for offers accepted in April, and sales reported in April were for offers accepted in March, etc. One very interesting note from realtor Barry Magee is that the percentage of sales reported in May (entry date=May) that actually was from an offer made in a month other than May (sold date=April, March, etc.) has changed. Here’s his graph:
I noticed that this has anti-correlated with how ‘hot’ the market is, which is reflected by the sales to active listing ratio (SAL). The following plots illustrate this:
What this implies is that when the market is ‘hot’, the diligence period, which is the time from the sold date to the entry date, is shorter. This makes sense, as in the hot days of 2016/2017 buyers were frequently giving offers without subjects and the ‘diligence period’ was actually the time that it took for the realtor to report it to the board (less than 5 days, more on this below). Indeed, if we plot the average diligence period for the months for which we have data the trend is clear, the diligence period is on the rise:
In the MVHC Facebook group, Phil Moore, former president of the REBGV, indicated the following:
Once a sale becomes firm ( all subjects are removed) it must be reported to the board within 5 days at the most. Staff immediately enter it into our mls computer as sold with the sales price. So each day sales are processed as sold but they have not been registered at the land titles office.
So if we assume that sales with less than a 6-day lag from the sold date to the entry date was a sale without subjects, the percentage of sales in that category has definitely changed as shown here:
If these assumptions are true, about 1/3 of the sales in spring 2018 were without subjects, and that number has changed to about 1/10 in spring 2019.
This also is reflected in the number of records for sales in 2018, where the periodicity is striking. Every Monday there was a spike in sales, as presumably the offers were presented after an open house from the weekend:
Contrast that to spring 2019:
What this says is that not only sale volume has changed in 2019, but also that on average, buyers are not buying immediately after open houses, they have subjects, and the diligence period is increasing.
Sales in 2019 have consistently reached multi-decade lows. May 2019 did have an increase in sales and sales to active listing ratio. The data suggests at least three reasons for this:
- Seasonally, May has larger sales than other months (see also our previous post on seasonality).
- Prices have dropped.
- Diligence period has increased
Next week spring is officially over, and summer begins. Historically there is a significant decrease in activity in Real Estate until the Fall. We’ll see what June, the last month of the spring season, ends up being. As of June 14, half way the month, the extrapolated sales to active listing ratio has again dropped to ~14% so far:
It’s still too early to tell what will happen in the second half of the year. I plotted how the US-correction from 2008 went vs the current Vancouver housing correction to get a sense of the timeline so far, this is the plot. We are approximately where the US cities were in January 2008 for their correction:
Will we follow course? Nah, it is different this time, right?