Numbers don't lie – especially about Canada's overheated housing sector. Increases in home prices in recent years have not been matched by underlying increases in fundamentals. The market is due for a severe correction.
One particularly telling statistic is housing investment as a share of GDP. This ratio has steadily climbed toward a record high over the past two years. It is now more than 7 per cent of GDP versus a 50-year average of 5.8 per cent and previous peaks of about 7.26 per cent in the late 1970s and 7.18 per cent in the late 1980s. After residential housing investment as a percentage of GDP peaked in the previous two cycles, the housing market crashed within a few years.
But while cyclical ups and downs in the housing market are to be expected, my forecast is not simply for a cyclical downturn. Rather, I see a long-term fall in home prices that will not be reversed when the economy picks up.
The key factor is demographics. House buying tends to follow the life cycle. Individuals accumulate assets when they are young and sell them as they get older and approach retirement. A country with a large chunk of its population in its working years will experience higher demand for housing – and therefore higher house prices – than countries with particularly young or especially old populations.
The evidence suggests that an increase in the proportion of people in their prime working years, between 20 and 64, has a positive effect on housing and real estate returns. The opposite is the case when the share of population that is younger than 20 or older than 64 increases.
One way to put a number on the demographic effect is to calculate the portion of the population that is not in their working years (people younger than 20 or older than 64) and divide this by the share of the population in their working years (between 20 and 64). A decrease in this ratio suggests that more people, relatively speaking, are in their home-buying years. An increase indicates the opposite, namely, that a smaller portion of the population, relatively speaking, is in their home-buying years.
The accompanying chart plots this ratio against house prices in Canada and in Toronto. There is a perfectly negative relationship – as the ratio has fallen, home prices have shot up. That is just what you would expect.
The demographic effect makes itself felt beyond the housing market. Households treat real estate like a savings instrument, so changes in population patterns have a similar effect on the bond market as they do on housing.
If we plotted the population ratio against bond prices, we would see a similar pattern as plotting the population ratio against house prices. As the population ratio has decreased over the last 20 years (indicating a growing portion of the population in their prime asset-gathering years), bond prices have risen. Since bond prices and bond yields move in opposite directions, yields have plummeted as bond prices have shot up.
Those low yields are reflected in low interest rates. And as we all know, low interest rates have helped to drive the housing market in Canada.
How much could home prices fall over the next few years? Every region is different, but I have developed a model of condo prices, based on Toronto condo rents, which uses investment-grade bond yields as discount rates.
My model estimates the condo prices that are, in theory, implied by current rents. The underlying assumption is that people should want the same cash flow from an investment in a condo as they would derive from putting the identical amount of money into a bond portfolio.
Assuming the rental market is properly priced, I calculate that the price of condos in Toronto should be, on average, about $100,000 less than they are now. As a result, one can expect about a 25 per cent decline in condo prices over the next few years. In the long run, the decline could be even more significant.
Of course, different regions may have different results, but I believe a perfect storm is coming in the housing market. Canada will experience significant "secular," or long-term, decline in house prices starting around 2015, when the population ratio is about to turn upward based on Statistics Canada projections.
The extent of this correction will surprise many people. Once the cyclical decline in housing is thought to be over and everyone prepares for an upturn in the housing market, the long-term impact of demographics will make itself felt and it will not be pretty.
George Athanassakos is a professor of finance and holds the Ben Graham Chair in Value Investing at the Richard Ivey School of Business, Western University. Read more of his Globe and Mail articles here.