Perhaps the hardest thing to forecast in the economy is turning points: where boom turns to bust, or vice versa. It is relatively easy to argue, for example, that an asset such as housing is overpriced or in a bubble, but much more challenging to predict when the bubble will burst – and still harder to make money on the bet. As John Maynard Keynes is attributed as remarking: “Markets can remain irrational longer than you can remain solvent.”
Indeed, the mere mention of the word ‘bubble’ is not without controversy. Believers in efficient market theory think bubbles are no more than an effervescent illusion, because everything is correctly priced. Eugene Fama – founder of the efficient market hypothesis – admitted in 2007 that “the word ‘bubble’ drives me nuts.” As behavioural economist Robert Shiller wrote in 2009, “you won’t find the word ‘bubble’ in most economics treatises or textbooks… the idea that bubbles exist has become so disreputable in much of the economics and finance profession that bringing them up in an economics seminar is like bringing up astrology to a group of astronomers.”
It is relatively easy to argue… that an asset such as housing is overpriced or in a bubble, but much more challenging to predict when the bubble will burst
Despite these caveats, let’s go ahead and make a prediction. Canadian housing is in a bubble, caused in large part by the existence of extremely low interest rates. The bubble has been assisted by government policy, valuations that boost prices, and a generally gullible media. Now, it’s about to deflate.
Canadian overvaluation
I am not alone in arguing that house prices in cities such as Toronto and Vancouver are looking a little frothy. Two typical ways to measure housing valuations are the price-to-rent ratio and the price-to-income ratio. Earlier this year The Economist wrote that Canadian house prices are overvalued by 78 percent in relation to rents (the highest in their survey of 18 countries), and 34 percent relative to income. An OECD report came up with over-valuations of 65 percent to rents and 30 percent to incomes.
The problem has in a way been the success of the Canadian banking system. In the years preceding the financial crisis, house prices followed a similar upwards trajectory as in the US. When the credit crunch hit, Canadian banks sailed through relatively unscathed. However, the central bank quickly lowered interest rates in concert with other countries, and kept them there to protect the fragile economy. It also extended mortgage periods to as long as 40 years, though that has since been scaled back to 25 years. The result was that house prices took a dip in 2008, but, unlike the US, quickly resumed their upward progress. Canadian housing prices used to track with US house prices, but now a gap of about 50 percent has opened up.
An accurate rating and appraisal system can act as a brake on prices by limiting mortgage availability – but this case may be different. In the 1990s, an automated program known as EMILI was created by the Canadian Mortgage and Housing Corporation (CMHC) as a quick way to determine whether a loan was at risk of default. The program was soon made available to banks for use in assessing mortgage applications.
The aim of the automated system was to make risk assessment quicker and easier. However, the program can be gamed in a number of ways – the easiest being to just enter false information (e.g. boost the square footage). As the Office of the Superintendent of Financial Institutions warned: “Automated models… have their drawback – primarily that they are driven by the sellers’ listings, which often inflates the value of the home.” As seen during the credit crunch, credit rating is a tricky game, where ‘quick and easy’ is no substitute for accuracy, and it might actually be worth visiting the property in question to get an idea of value.
Driven by predictions
The market has also been driven higher by predictions that the market will go higher, or at worst suffer a so-called soft landing. These predictions are supplied by banks and real estate organisations, and regurgitated by major media outlets.
A first question that should be asked about any prediction is whether the forecaster has an interest in the outcome. Financial firms and real estate companies have a lot to lose from a housing bust, so any of their statements should be taken with a pinch of salt (why on Earth would they predict a crash?). For alternative viewpoints you need to go to places like Capital Economics, which for a couple of years now has been predicting a 25 percent slump.
As I argued in Economyths, fluctuations in housing markets are similar to long-term weather patterns such as El Niño. Both are driven by complex, global, interlinked factors. Both are very hard to predict using even the most sophisticated models (in fact simple models often do a better job). And while you might suspect that you are in an El Niño – that is, an overvalued housing market – it is less easy to know when the opposite phase – La Niña, or a fall in prices – will kick in.
The Canadian housing market has defied gravity for years. What will it take to pull it down? It may be that the process is already under way. One clue when analysing the dynamics of a system, be it the weather or the economy, is to look for properties which define a characteristic timescale. In the economy, such numbers are interest rates or inflation, which have units of 1/time (e.g. percent per year). A period with low interest rates and low inflation will often be associated with a lower sense of urgency and correspondingly slower timescales – leisurely booms which extend well past their best-by dates, followed by gradual declines.
As Shiller remarked last year: “I worry that what is happening in Canada is kind of a slow-motion version of what happened in the US.” Instead of a crash, expect a gentle sag. At least until interest rates go up, as they eventually will (an even safer prediction).
