How long can the housing madness last?

Back in 2005 I bought a house. I knew then, as I know now, that I have no ability to predict the future. But that is not the same as lacking expectations. One of mine was that rising home prices in Toronto's residential real estate market would not continue for much longer, and that in the not-too-distant future my house would be worth less than I paid for it. My thinking was dominated by a belief that housing markets were cyclical, and Toronto's boom was getting long in the tooth.

I was dead wrong, of course. The Teranet-National Bank House Price Index (coincidentally set at 100 in July 2005) now sits at 133.5 in the Greater Toronto Area, meaning prices have increased on average 33.5% in the past six years. The Canada-wide index is up even more. Canada is one of a few developed countries in which housing markets weathered the recession and continue to advance. (Others include Australia, Belgium, Netherlands, Sweden and Switzerland.) Elsewhere, such as the U.S., Britain, Spain and Ireland, prices are in retreat. Call me stubborn, but I still believe Canada's housing market rests on a precarious foundation and will not escape a correction, anymore than did America's or Britain's.

How long can this madness continue? Philippe Bracke, of the London School of Economics, recently shed light on that question. Bracke noted that home prices have exhibited cyclical behaviour for centuries. "Available historical records show that this recurring sequence of house price expansions and contractions has been a constant feature of industrialized economies at least since the 17th century," he writes in a recent working paper. Bracke reviewed 40 years of housing data from 19 countries to determine how long these cycles typically last.

Of note, Bracke's data shows that the mean duration of an upturn is about 28 quarters (i.e., seven years), while downturns last 18.4 quarters (about four and a half years). Readers will note that this suggests upturns tend to last longer than downturns. However, the most recent, largely synchronized boom (the one Canada's still in) has been unusually long; when Bracke excluded it from the data this difference largely evaporates.

The data also revealed that the longer a boom continues, the more likely it will end. This may seem blazingly obvious, but in fact it isn't: that doesn't happen with downturns. Bracke hypothesizes that speculation and overbuilding help explain why booms become increasingly unsteady as they persist. Citing theories of boom-bust cycles in home prices, Bracke writes: "As expansions get longer, they are increasingly likely to terminate, signaling a progressively unsustainable departure from fundamental price valuations."

Although Bracke primarily studied duration of booms and busts, he also considered amplitude—that is, the magnitude of price changes. Here the data is somewhat encouraging for Canada. Prices here since 1998 have risen 86.9%, which is quite extraordinary by historical standards. Still, it's nothing compared to 286% increases in Ireland between 1994 and 2006, or the 200% witnessed in Norway between 1993 and 2007. Then again, one should note that America's famous housing madness between 1995 and 2006 saw an increase of just 64%.

In Canada prices peaked in 1976, troughed in 1985, peaked again in 1989, troughed in 1992, peaked in 1994 and troughed in 1998. The run-up since then is now 13 years old (about double the mean duration of run-ups in developed countries over the past 40 years) and still going strong. There's no magical rule that because something has gone on for an atypically long time that it must reverse. However, of the 19 countries Bracke studied, "no nation has ever lived through a perennial house price expansion or contraction." Statistically speaking, I have to be right one of these days.