Reading Tea Leaves – Predicting Canadian Recessions Using Financial Variables February 5, 2008
Posted by radley77 in recession risks.Tags: bonds, commercial paper, recession, yield spreads
trackback
One of the tools that the Bank of Canada uses for forecasting probability of recessions is the spread between Government of Canada 10 year bond yields and 90 day commercial paper. This is a leading indicator of slower economic growth. The benefit of using a tool like this, is that if recessions can be successfully predicted in advance, monetary policy can be adjusted accordingly.
The paper “Predicting Canadian Recessions Using Financial Variables: A Probit Approach” concludes to say:
“Results in the paper show that, in comparison to other financial variables, the spread between Canadian long bonds and the 90-day commercial paper rate is best at predicting recessions in Canada.”
The following graph shows that the spread has been the largest since the 1990’s recession for about a year:
The past two recessions have marked peaks in the housing cycle in Calgary. As carrying costs are the highest since previous housing bubbles (see this post), it is useful to estimate the probability of a recession as a tool in forecasting the peak in the current housing cycle.
I am optimistic that Canada will avoid a recession at this point. However, I estimate that the next period will be the slowest economic growth that has been seen in more than a decade. I estimate that the probabilities of a Canadian recession are about the chance of flipping a coin three times and having all heads. Currently, Global Insight has estimated the probability of recession at 25% in Canada, with most other instutions forecasting less risk than that.
For those that are highly leveraged and unable to cope with an economic shock, there may be difficult times ahead…
Interesting analytics Radley. I would say that the central banks actually have no reliable statistics that can predict the next recession. Our friends at the Fed basically just proved that. Until a few months ago, the “subprime” problem was “contained”, and the credit markets would be solved in a few weeks.
The reason that they couldn’t predict this recession is that it is the result of a different set of factors than previous recessions, namely, the unwinding of the credit bubble. My guess is that they still don’t even believe there WAS a credit bubble. Past recessions were caused by different factors, and hence old indicators will not be to useful in forecasting the next recession, in my opinion
My prediction is that the next Canadian recession will be due to weakening global demand (dropping demand for many commodities), along with the unwinding of the world-wide credit bubble, and the massive level of debt that Canadian citizens have put on themselves. Once the public decides to start saving and reduce their debts (ie consume less), the economy is go into recession. The last recession was driven by an equity market bubble bursting (Nasdaq), but was very short, due to the healthy balance sheet of US consumers, who kept on borrowing and spending. This time, they can’t borrow any more, and so this recession is consumer driven, not corporate driven.
In my humble opinion, most Canadians (including politicians) are clueless as to both the roots and nature of the credit bubble, as well as the implications of that bubble unwinding. The masses think that Calgary home prices are based on “demand”, without understanding that demand is largely a function of credit availability.
Longterm investors will settle for lower yields now if they think rates and the economy are going even lower in the future. Traditionally, it means that they are betting that this is their last chance to lock in rates before the bottom falls out.
If you look at the 1991 case, by buying a long bond, you would have locked in at a higher interest rate and missed the recession. Wahoo!
This would have been more beneficial than buying the 90 day commercial paper.
As we are in a low interest rate environment, I would think that the spread is still a sign of worrisome things to come.
As inversion of the spread between the two happen rarely, I would think that it something important not to ignore.
Alas, there are no silver bullets in predicting recessions.
If you are interested, there is a similar paper by the Federal Reserve called “The Yield Curve as a Predictor of U.S. Recessions”:
http://www.newyorkfed.org/research/current_issues/ci2-7.pdf
By the way, I totally agree with your thoughts on the credit bubble. I remember a few years ago when I read that Americans on average had a negative savings rate. I assumed it was only a matter of time before credit markets would started unwinding. That is one of the flaws of having a bank like the BoC that’s mandate is to target inflation… Other aspects of the economy like asset bubbles and debt markets have a way of creating problems and there is a very limited way in which the central banks can intervene.
By the way here is a historical yield curve calculator.
http://www.smartmoney.com/onebond/index.cfm?story=yieldcurve
See if you can find the dot-com bubble, Katrina and subprime indicators…