Earlier this year Dr. Moshe Milevsky updated his seminal research on the advantage of variable interest rates. The update further reinforced the benefits of variable rates illustrated in his original 2001 study.
The findings, which we wrote about in April, do not tell the whole story however.
To gain a more in-depth understanding of his research, we had the privilege and pleasure to speak with Professor Milevsky himself. Our discussion revealed several notable insights which we’ve compiled below in the first of our two-part interview.
On whether his latest research suggests variable rates are the best choice for most people.
Dr. Milevsky: One of the things that bother me most about the debate with mortgages is that it is all or nothing. It reminds me of 30 years ago when people used to ask: “should I buy this stock or should I buy that stock?” Nowadays people use diversified portfolios that take account of one’s attitude towards financial uncertainty and one’s willingness to trade-off risk and return. There is no one-size-fits-all-solution. That type of comprehensive and holistic thinking is what I’d like to see applied to debt management.
On how mortgage planners should apply this research when advising clients:
Dr. Milevsky: The purpose of the study is not to give financial advice; it’s to document the historical record. I like to think of these studies as being parallel to the asset management and investment studies which document that stock funds beat bond funds in the long-run, or that real estate beats inflation, in the long-run, etc.
On whether Canada’s historically low interest rates affect his conclusions:
Dr. Milevsky: It’s more about implementing a balance sheet approach to debt management. Mortal minds cannot predict the credit markets. We don’t know what the yield curve will do next month. The sensible approach is therefore to have some fixed-rate debt, some variable-rate debt, some long-term and some-short term. Assets are commonly diversified, so why not debt? To me, the question in mortgage financing should be: “How much of my mortgage should be closed and fixed versus open and variable?” as opposed to the more common – but erroneous – which is the best mortgage today?
The wrong way to select a mortgage is for a 27 year old mortgage broker to say “I’m smarter than the Bank of Canada or I’m smarter than the quants who manage hedge funds.” You can’t outguess the market.
On when to go fixed:
Dr. Milevsky: It depends on the individual. If someone has a 99% loan-to-value I’d tell them to go fixed. With 50% equity I’d tell them to take a variable-rate mortgage and have their payments fixed if necessary. You need to see if they can afford the swings by looking at the totality of assets on their balance sheet.
On why his study is based on 10 and 15-year amortizations instead of 25:
Dr. Milevsky: Ten and 15 year amortizations were chosen partly because they represent the halfway mark for the typical Canadian homeowner. This timeframe also simulates the remaining amortization of people who are refinancing. We didn’t want to signal that the study applies only to people who are buying a new home.
Another point is that a 25 year amortization would make it more difficult to obtain accurate results due to a lack of available periods. There are not that many non-overlapping 25 year periods in Canadian financial history.
On whether locking in a variable-rate mortgage to a fixed rate contradicts Dr. Milevsky’s findings:
Dr. Milevsky: People’s attitude to risk does change. It’s not in your DNA that you are permanently risk adverse or permanently risk tolerant. Risk aversion can go up over time. After a period of time it may very well make sense for a certain individual to lock in. One example in which circumstances might change is if rates go down dramatically, so that it is physically impossible for them to decline any further. Another example is if you extract some equity from your home (perhaps to renovate) and are back to a high ratio mortgage. These are changes in financial circumstances that might justify locking (more) in.