, , , , ,

In many recent articles, the Globe and Mail has been reporting on the facts and consequences of debt in Canada. There are some familiar-but-alarming figures: debt is 163% of income on average and rising; Canadian debt totals $1.8 trillion; there is no such thing as good debt (well that is debated: owning a home may or may not be, university education usually gets a thumbs up); using even one payday loan is very expensive; policy makers are or should be spending a lot of time thinking about monetary, fiscal and macro-prudential policies. On the individual level, there are lots of stories showing how “too much” debt makes life stressful.

The stories include a little bit of good news (one-third of Canadians have no debt) and the scary stories sometimes reveal good decisions. Since lots of people lose their job even in good times, you should have an emergency fund (e.g. 3-6 months of income just in case). Unless you are fantastically wealthy, you should pay off credit card balances as soon as possible. You should match current consumption to current income to prevent future problems (e.g. phone calls from credit collection agencies or bankruptcy).

Mostly, I think that the articles miss an important point. Like most media reporting, scary personal stories attract attention. But, as I noted in a recent posting on problem solving and real estate, an excessively narrow focus does not necessarily identify the real problem.

The reason why people worry about other people using debt to pay for real estate is the possibility of contagion. Individuals can get into trouble for their own reasons but the recent U.S. experience shows that events in the real estate market can affect mortgage debt which can affect people and financial markets world-wide.

So, the important and overlooked question is: does the current state of real estate markets in Canada represent a systemic risk?   I have a two part answer. This first part notes how bad the data is. A second part uses logic plus the available data to suggest that any problem is not contagious. (There really should be a third part of whether a crash of the market in one location is contagious nationally but, for now, too many people fixate on the idea of “Canadian real estate market”.)

This question is hard to answer even with the benefit of advanced methods and lots of data. So, the first comment must be to repeat an idea mentioned in the Globe and Mail articles: the easily available data is poor. As a researcher, what I would really like to know is not the average ratio of debt to income but the fraction of individuals who would have trouble repaying if something happened. The average is largely irrelevant since the riskiness of the situation depends on whether everybody has a 163% ratio or whether 90% of Canadians have 25% ratio and 10 percent of individuals (concentrated in Toronto and Vancouver and not Saskatoon and not Moncton, to make things scarier) have a 1400% ratio (yes, these numbers produce the same average overall). The degree of worry also depends on what “thing” happens: an “interest rate increase” does not mean that everybody starts paying 10% interest on their mortgage immediately.

Concern also varies with how debt is being used. One of the features of the US Crash in ’07, that is now being forgotten, is the stories of a person’s home being used as an ATM to fund unsustainable consumption. These stories reveal a mechanism which links real estate markets to a systemic economic risk: if house prices stop increasing then people stop borrowing for consumption and the macroeconomic downturn is expected. While the logic is valid, the premise has not been backed up by evidence: I have not heard such stories in Canada (although the growth in home equity loans of credit (HELOCs) is a suggestion worth pursuing).

Without good evidence, anybody can say nearly anything (and you can usually find somebody who does). As a researcher, I find debates amongst advocates frustrating because debates using conflicting anecdotes can go in circles endlessly. The fact that so many people can talk with so much assurance for such a long time, without proposing a convincing alternative decision that they think is better, is also frustrating. Since you may know exactly what you need before it is used, investing in good data collection processes is something which has great long term benefits for society.