Saturday, January 06, 2007

Rental Vacancy Rates and their relation to Price Growth

Rental accomadations can be hard to come by these days. Fortunately, my wife and I own a place we call home and we like it quite a bit, but I have heard that it can be difficult to find a suitable place to live these days. I was thinking about this and the crazy real estate market of the Greater Vancouver area over the past few years and I wanted to do some analysis on this.

What did I find?

Well, first off, I don't want anyone to draw a cause and effect relationship between these two variables (real estate price growth and vacancy rates) but there is, quite obviously an inverse correlation between price changes and rental vacancy rates. We can speculate about the cause and effect relationship.

Here is the chart (based on data from UBC's Sauder School

What this data tells me is that when year over year price growth in real estate is high, vacancy rates tend to be low and when price growth is negative or low, vacancy rates are high. For the math fans out there the correlation coefficient is -0.69.

What do you think? Why does this happen? Is it because of the diminishing returns for investors? Transactional overhang? People not buying so it drives demand for rentals? How could we draw a cause and effect relationship?


Uncertain Buyer said...

Looks like you are off to a good start. I will toss a link on my Blog

You might want to find a way to make those Graphs link to a larger image. With my resolution they look small and are hard to read.

mohican said...

Thanks for the feedback. I'm new to all of the blogger features and tricks. I think I've fixed that problem now.

solipsist said...

Welcome to the blogosphere mohican. Your blog promises to be a welcome look into the ephemera of the RE market. That graph from Sauder is fascinating.

I will put up a link to your site at vancouver (un)real estate right away.

ADWeaver said...

Hi Mohican,

nice to see another quality RE blog that I can add to my weekly reading.

I think YLTWNBoomerang hit it in a post he made at VHB (where I found your link).

YLTWNBoomerang said...
I've lived in Yaletown the past 5 years and have noticed:

1. In the newer buildings that have completed in the past 2 years there are a lot of unoccupied parking stalls yet always ads on the peg-boards looking to rent stalls (my building and friend's buildings too). This tells me there may be a lot of unoccupied investment units left vacant to preserve newness and allow for an easy flip otherwise the stalls would be rented.

2. In winter midweek, around dinner time, have a look around at the buildings and you will not see many lights on. Why aren't people eating/preparing dinner? Perhaps they are eating out? Check out Yaletown restaurants and they're pretty empty during the week on average and only get busy on weekends when non-locals head into town for a weekend meal out!

3. The above reinforce that there are a lot of vacant investment properties. I then liken a property to equity in a company and rent to a dividend. I'm happy not receive dividends as my capital increases 10-20%/year however when that drops to 6-9% (as per CMHC), I have to think whether I should rent it out (views poorly, use degrades value) or dump it and invest my gains in some other investment.

Now suppose CMHC is wrong and housing goes up only 0-3%...I can do better with almost any investment and even get a better return with an ING 3.5% savings account. Of course there are capital gains on this income, but if the investor were offshore to begin with it makes no difference...

Just my observations

1:37 PM, January 06, 2007

So, as we see steep rises in property values, investors are more willing to keep their investment properties vacant. Once things flatten out, or start to decline, any investors holding properties will be far more inclined to rent or sell.

In theory, if there is a higher percentage of properties being held as investments now (which I think there are) than there were before the declines that we've seen in the past (82, 95), that aspect of the market should exert a greater force on falling prices this time around.

Great graph. Thx.

rentah said...

Thanks for the graphs, mohican.

Wu'kong said...

Interesting -- have you run the test to see how strong the correlation is?

I'm feeling too lazy right now to do anything but suggest it ...

mohican said...

Thanks for the comments.

There is a lot of RE data out there just begging to be analyzed. I'll see what other interesting correlations and trends I can draw out of some of the Sauder data.

the pope said...

Thanks for the new blog and analysis mohican, I'll definately be posting a link to your site on

I wonder how much of this effect is related to the froth of buying and selling? I know of a few houses standing empty while the new owner tries to sell their old house. Older apartment conversions to condos would also temporarily drive down supply creating lower vacancy rates.

Looking forward to more graphs!

Uncertain Buyer said...

Is there any way of graphing the US market and Canadian market to see if there is a common trend? I have heard that the US is about 6 months ahead of Canada in housing trends. It would be interesting to see this. It might be useful in looking ahead to see where we are going.

mohican said...

I think that there could be a few causes for the low vacancy rate.

I have a feeling that the as the number of sales rises in a market the number of homes sitting empty rises as people wait to move in, renovations take place, people decide what to do, etc. I call this transactional overhang and I think it could represent a significant portion of why there are less units available to rent. I'd like to show this on a graph and I am trying to find solid data to use.

The other reason is likely the result of owners deciding to stop renting and sell or as was mentioned - condo conversions. Converting a rental property into a for sale unit. I know of several of these type of situations here in the Surrey, Langley, White Rock area. This would be difficult to prove as I know of no repository for data that tracks conversions.

Rob Chipman said...


I'm linking to you as well, under real estate blogs. I for one will draw a cause and effect relationship between rents, vacancies and values. I don't think the effect is always as straighforward as we'd like, but its there.

Wu'kong said...

A quick suggestion: since your origins and scales are different for your axes, perhaps you should try and normalize your data before looking for graphical correlations.

mohican said...


I actually didn't run a graphical correlation analysis. I ran the analysis on the original data arrays which I imported to MS Excel from the Sauder data.

Thanks for the feedback. I'm a little rusty on my Excel but its all coming back to me now. Any other feedback is appreciated.