Are vacancy rates a leading indicator of house prices?
DECEMBER 16, 2011
Here's a great reader comment that I've actually recently been doing some work on:
"Vacancies were flat in Toronto until 1990, and flat in Miami until 2007, and I think I saw a few more examples (haven't gone after Vancouver's past data yet). But you would think that if there was oversupply as part of a boom, that vacancies would go up before the crash in prices hit."
It's a great point. In most large cities in Canada, vacancy rates are benign (Toronto, Vancouver, Montreal, Ottawa) or are falling from elevated levels (Calgary, Edmonton). It's also a great point that in most large US cities, vacancy rates really took off AFTER that bust was well underway. The rate was 2.9% in Miami and 2.7% in San Diego in 2005. Today Miami's vacancy rate is 8.4% while San Diego's is now 6.7%.
Before discussing why I think vacancy rates in Canadian cities are as much a lagging indicator as anything else, let's review some extremely pertinent information that frames this discussion.
In the past decade, all provinces have built significantly more residential dwelling than demographic demand would warrant...and at a time when house prices have been very strong....a bit of a paradox if we act like traditional economists and assume that consumer behaviour remains constant over time.
Indeed, we've seen the population/residential dwelling ratio fall to new all-time lows. While this is certainly not a new story (it's been falling since the 60s), it will reach a concrete lower boundary.
Conventional thinking would assume that if you build more homes than people need, it must pressure house prices and/or raise vacancy rates. But once again, conventional thinking fails to take into account the massive shift in consumer behaviour over time.
There are several factors that I would argue play a significant, if unquantifiable role in hiding this mystery ‘inventory’:
1) Speculation: We’ve seen widespread speculation in markets like Toronto (particularly the condo market where 25% of all new units are being sold to flippers and 60-80% to cash flow negative investors) and Vancouver (where a municipal election candidate recently wrote of her experience canvassing in areas where virtually every house on a block is unoccupied).
2) The willingness of consumers to hold vacation homes that for all intents and purposes could be residential dwellings: In its 2009 recreational property report, Royal LePage Real Estate Services noted that, “64 per cent of Canadians view cottage ownership as a sound investment. To pursue their dream of buying a recreational property, 55 per cent of Canadians would be willing to make compromises with regards to their financial or lifestyle choices, such as purchasing a property with family and friends, renting out their cottage, making a cottage their primary residence, buying a fixer-upper, or moving into a smaller principal home in the city.”
While people are quite happy owning multiple homes during a housing boom when bullish sentiment is rampant and economic growth is strong, these properties tend to hit the market first when the boom ends or the economy sours. I'm currently digging into tax data to see how I can further quantify this, but the initial evidence suggests that the growth in vacation homes has been very significant over the past decade.
3) Change in household formation patterns: During housing booms associated with cheap and readily-available credit, there is an increase in households that are formed when children move directly from their parents’ home into a residence of their own, without renting. This is facilitated by the loosening of underwriting standards, a trend we’ve noted in Canada. Yet all our data strongly suggests that the economy is not nearly as sound as it appears on the surface. And when the recession finally arrives, these same individuals may find it difficult to make ends meet, opting instead to move back home. One needs to look no further than the US experience to see how household formation patterns accelerate during a boom and decelerate during a bust, in part because of this very dynamic.
The main point is that vacancy rates can remain low even while starts outpace demographic demand due to changing mass psychology. And while the vacancy rate may appear benign, other data points suggest that a potentially significant inventory overhang will likely appear if the real estate market begins to melt and the economy sours with it.
Cheers
Ben
Posted in:
- |
- Tagged:
- canada,
- vacancy rate
19 Comments
What is this crazy stuff? Welcome back and great post like always, You were badly missed.
Now a question for you? When do you expect vacancies rates to go up here, one thing for sure if vacancies rates do not go up, pprices are never coming down.
I was surprised by the Ontario chart. I thought with the cranes everywhere in TO, that housing starts would be way above household formations for the last few years but the opposite is true, am I missing something?
Great post, as always.
I suspect that if housing supply could be charted as available bedrooms per capita the oversupply would be even more dramatic. 3 people using 2 beds in a 6 bedroom, 5 bath mansion kind of thing.
What is even more puzzling to me regarding vacancy rates, is the large number of illegal basement apartments that are housing thousands of people and yet are largely unaccounted for in the determination of available rental stock.
Without these illegal apartments, the rental vacancy rate would be 0%, with long waiting lists in every major Canadian city.
CMHC rental housing survey data does not consider any basement apartments (even legal ones) as part of the available rental stock. In fact, any apartment in a building with less than three rental units is excluded, and units that have been vacant for less than three months seem also to be not actually considered "vacant". All this from the methodology description on the CMHC's website.
I would not be surprised if the actual rental vacancy rate is double the number reported by the CMHC.
CMHC Toronto vacancy rates updated. http://i41.tinypic.com/wmji13.png
I don't agree with Miami/Toronto vacancy rate comparison because they both had a robust amount of immigration, however the difference is Miami immigrants were primarily lower income Hispanic/Latin-Americans as opposed to Toronto's immigrants who are primarily Asians, of which the majority are i) Filipino woman who work in healthcare (government jobs provided) ii) wealthy Chinese families/students/investors.
There is a big difference between immigrants who migrate in order to seek a living as opposed to those who migrate with a job placement, to study or to buy/invest a home as a safe haven.
With that said, here is the 2012 immigration targets (this does not include foreign workers and students) http://www.cic.gc.ca/english/department/media/notices/notice-levels2012.asp
It doesn't take a genius to figure out the loophole here—when all you need is one family/economic immigrant who can now employ friends using the fast-track foreign worker program that allows immigrants to get in the country within 6-12 months. This is one of many loopholes in the system and I find them reading immigration blogs where they discuss ways to get in to Canada.
The immigration system is a complete disaster.
Greg....really want to chat about the source of some of your graphs. Can you email me? Ben@theeconomicanalyst.com
Cheers
That was a quick request fulfillment :)
Thanks Ben.
You're welcome!
In Vancouver, Burnaby, and New Westminster suites in detached houses are easily half the rental stock.
Vacancy rate is just one of the many indicators that has to be used with understanding, insight and care.
In good economic times, where jobs are plentiful, a single person can rent a one bedroom or even a two bedroom and have the financial means to support it. In bad economic times, this same person may share it with another person.
When the local economy is good, people migrate to this local to have a job, thus rent a home or buy a home. When things turns sour, people migrate OUT this local, thus increase the vacancy rate.
Vacancy rate is kind of like the P/E ratio of a cyclical stock. When in the recession, the cyclical stocks usually sport high P/E ratios of 40 to 50. To an inexperienced investor, he or she would think this stock is ridiculously over-priced!!! But to an value investor, this is actually the time to buy these cyclical stocks, because they are cheap!!!
Currently the low vacancy rate in TO is used as one of the final remaining tricks to induce people to buy into the real estate. So buyers be aware.
Off topic, but a nice blast from the past, highly relevant to the general theme.
Calculated Risk in 2006 as the first cracks were appearing in the US housing.
http://tinyurl.com/c6vurzc
First, the bland reassuring pablum from an "expert":
There's just too strong an economy and too much job growth for much other than the "soft landing" Husing and other economists have been predicting for the end of the five-year housing boom.
Calculated Risk begs to differ:
As the housing bubble unwinds, housing related employment will fall; and fall dramatically in areas like the Inland Empire. The more an area is dependent on housing, the larger the negative impact on the local economy will be.
So I think some pundits have it backwards: Instead of a strong local economy keeping housing afloat, I think the bursting housing bubble will significantly impact housing dependent local economies.
Please note that the dangerously high share of direct construction jobs hit just over 10% in the epicenter of California's bubble, the Inland Empire area.
What is it in Canada's hot-spots again? Ben, I know you did a post or several on the topic, can't seem to find it now. Canada as a whole is 7.6% compared to the pre-bust US 5.5%. Vancouver especially is insanely high, like 20%-ish or so?
Sorry, As the housing bubble unwinds, housing related employment will fall; and fall dramatically in areas like the Inland Empire. The more an area is dependent on housing, the larger the negative impact on the local economy will be.
So I think some pundits have it backwards: Instead of a strong local economy keeping housing afloat, I think the bursting housing bubble will significantly impact housing dependent local economies."
should have been bolded as it is Calculated Risk's summation, not my words.
My bad.
Great interview yesterday, Ben. If anyone has not heard this Talkdigital radio interview discussion and the insights Ben offered on BC real estate you might want to tune into the podcast here...Ben's portion starts at 19:05 and runs ten minutes or so.
http://talkdigitalnetwork.com/2011/12/this-week-in-money-5/?utm_source=w...
The interview from yesterday is on a program called "This Week in Money" and it is from Howestreet.com
Glad you enjoyed it!
While I agree that a rising vacancy rate is a lagging indicator (we only know of its factual existance after an R/E correction has already happened), I think we need to appreciate that this is also a reasonable predictive tool if you feel sure a recession is coming.
In other words, when we know that a housing market has run out of steam and a credit bubble has reached an end we can also surmise a recession might follow and therefore that vacancies will rise. This, in fact, is how so many economists leave little doubt that house prices will be magnified on the way down as the set of circumstances that fed into the price appreciation are inverted and the process begins to run in reverse.
You do not even need rising interest rates to kill a housing bubble though, only a set of circumstances that shows the price/rent ratios are sweked, there are not significant numbers of entry level buyers and last, that applications for credit and new home construction are falling.
As Ben has pointed out, one outcome of the bursting of the US housing bubble was sharply rising vacancy rates. There were a number of reasons, the most important of which was an increase in the unemployment rate.
The simple reason is that people who are financially stressed and feeling insecure about the future and the economy tend to cohabit again and band together. There is more than enough ample living space available at this time in most cities. We just do not use it efficiently (yet).
Actually, there is a good probability that Toronto has a real surplus where condos are concerned.
My point here though, is that as we know the market has topped in Canada for home buying and the credit orgy has reached an end, we can easily see what lies ahead. I do not need to wait for the actual statistics on rising vacancies to know that housing prices will see significant price declines in some regions.
The indicator lags....but it is wholly predictable.
Hi Ben,
I enjoy your posts. I agree with most of what is written about overvalued RE being driven by easy access to credit. I also believe we are in for a significant correction.
I was interested in one chart above, residents per dwelling, and the possible conclusions to be drawn. It would appear that the reduction in residents per dwelling in the last 15 years (particilarly in Quebec) is a significant driver of increased demand for housing. No stats to back this up, but I would hazard a guess that many of these smaller households belong to:
1. Young singles. People are getting married at a much later age. As such, young professionals are accumulating wealth and can afford to buy a studio or one bedroom condo in most markets across the country.
2. Middle-age singles. For the past 15 years divorce rates have been relatively constant at roughly 2 per 1000 people per year. Annecdotally, in the past many divorcees would move into rental accomodations becuase there wasn't a supply of smaller, less expensive homes.
3. Seniors. Life expectancy continued to rise gradually and steadily over the past 15 years. http://www.statcan.gc.ca/pub/82-229-x/2009001/c-g/c-g-1/ch1_graph2.0-eng.... This has to impact the housing market. Annecdotally, I know many seniors who have downsized to condos (increasing housing demand). As seniors live longer and healthier lives, if they aren't downsizing that means fewer homes are coming on the market (decreasing housing supply).
Individually, these demographic changes may not have much impact on the housing market. Collectively, I wonder what the impact has been? It strikes me that in all the cases above, the multi-unit market has accurately met demand by offering smaller places with potentially less upkeep. Couple the condo boom with low interest rates and we find a significant change to demand where individuals choose to buy rather than rent. In all cases above, the lower price point of a condo means affordability ratios are likely manageable.
Is this worthy of further investigation?
In Alberta most condo townhouses and apartments lost around 40 to 50 percent back in 2007 and into 2008. Most haven't gained back one iota since then. That's what explains the chart of Alberta.