Canada's Housing Bubble

Sep 19, 2011 Ben Rabidoux theeconomicanalyst.com

Interesting insights from CREA:

Those awaiting part two of our series looking at the economies of various Canadian cities and which ones are most at risk from a housing correction will have to wait until tomorrow to see the craziness that is Western Canada.

For now, we have to look at the most recent data release from CREA (the Canadian Real Estate Association), which contains two flashing lights that may be indicative of a market peak.  Time will tell if these data points are the proverbial 'canary in the coal mine', but they are certainly not good news.

Before looking at that release, it is worth revisiting an important indicator that I believe may signal a turning point in the market.  The change in mortgage debt as a percentage of GDP should, theoretically, serve as a proxy for mortgage credit demand.  As Steve Keen likes to point out, people don't buy houses, people with mortgages buy houses.  And I have repeatedly warned, supply/demand dynamics can take a back seat to credit availability and credit demand.  You can have a 'balanced' market that is rising or falling precipitously on the back rising or falling credit availability.  That's a reality I explored in a recent post.  The bottom line is that the year-over-year change in mortgage debt as a percentage of GDP is now at levels not seen in a decade.  I suggested at the time that if this persists, we should begin to see house price weakness within a couple quarters:

With consumer credit demand having slowed to its lowest point in nearly two decades, there's rising evidence that peak credit may be near.  The lee side of that credit mountain looks vastly different than the way up.

Shifting our focus back to CREA's press release, here's their monthly commentary, which is always superficial and next to useless:

For a second consecutive month, national home sales activity held steady in August 2011 when compared to the previous month.

Among major urban centres, Toronto and Ottawa posted a monthly increase in activity while Calgary, Montreal and Vancouver saw activity decline slightly.

“The housing market in Canada remained on a firm footing in August when compared to volatile financial markets,” said Gary Morse, CREA President. “Through their actions, homebuyers are showing that they remain confident about the stability of the Canadian housing market, and recognize that the continuation of low interest rates represents an excellent opportunity to buy their first home or trade up.”

Actual (not seasonally adjusted) sales activity came in 15.8 per cent above national levels reported one year earlier. This was the largest year-over-year increase since last April, but largely reflects weakened activity one year ago.

This is not the sort of data that sounds indicative of a market peak.  But buried in the report are two nagging indicators that price weakness may be on the way...

1)  Diverging supply-demand lines:

Most commentators tend to look only at the supply-demand balance in forecasting house price movements.  I don't dismiss this as supply and demand do drive house prices when one significantly lags the other.  It should be remembered, however, that this is only one possible way by which house prices can come under pressure.  I recently discussed this exact concept.

With that in mind, note the supply-demand graph in the CREA report.  The top blue line represents new listings (supply), while the bottom red line represents sales (demand):

Note that for the first time since late 2008, these lines appear to be diverging:

Whether this proves to be a fleeting trend or the start of a sustained period of supply-demand imbalance is a crucial question that we will monitor.  At the very least, expect the months of inventory and sales-to-listing ratio to deteriorate over the next few months.

2)  Unusually large drop in average house price from Q2 to Q3

Perhaps the bigger head-scratcher is the unusually large drop in house prices between Q2 and Q3.  It looks as if Q3 is on track to see average resale prices fall 6% over Q2.  You can see the drop quite clearly at the very end of this chart from CREA:

Folks who read this site regularly will know that I often vent frustration at the use of quarter-over-quarter or month-over-month comparisons of house prices as they can be highly influenced by seasonality.  In this case, a decline in prices from the second to the third quarter of the year is very common.  It occurs more years than not (60% of the time to be precise).  What is unusual, however, is the depth of the drop.  If the August numbers hold through September it will mean a quarter-over-quarter decline of 6.5%.  And given that September typically sees lower prices than August, there is a chance that the final quarterly decline may be even greater.  As it stands, the only year that matches the current pace of decline is 2008, right as the financial crisis was beginning to grip the world.

I can't tell you exactly what this means or whether this is the start of sustained price weakness, but I know one thing:  The bulls will have a hard time putting a positive spin on this one.

Cheers,

Ben

ADDED:  Those interested in reading what RBC had to say on the CREA data release can read this report, hot off the press (as of 4pm).  I can save you the reading.  Three words:  Moderation, no crash.

Meanwhile, TD has added their own take on the latest data.  Their bottom line:  Expect a 10% peak-to-trough decline.

We'll see...