Predicting Interest Rates: Futile.

mortgage-risk Why do economists make interest rate predictions to the nearest tenth of a percent?  To prove they have a sense of humour of course. 

As the New York Times says, economists are paid to guess wrong.  In July 2007, for example, 13 of 13 major Canadian bond dealers predicted interest rates would increase in late 2007.  One month later 11 of 13 said rates would remain unchanged!  A month or so after, economists were actually calling for the BoC to lower rates! 

Things can change that fast, largely because the inputs in interest rate forecasting are infinite and random.  No one can properly process and weigh all those variables.  Indeed, research consistently shows that professional economists continually guess wrong over time. (See the Cleveland Fed Study, Dowling College Study, FMA Study, St. Louis Fed Study, etc.) 

In short, you might as well flip a coin than rely on an “expert’s” opinion of rate direction. 

But if no one knows where rates are going, how do you know whether to get a fixed or variable mortgage? 

Well, try asking that question of your mortgage planner.  The best professional mortgage planners will analyze your financial profile and risk tolerance, apply the available research, and then suggest the best course of action irrespective of rate direction.

We may not know where rates will be 12 months out, but there’s a few tricks of the trade we can still employ to save you interest.

  1. When trying to decide whether to go variable or fixed, I usually base my decision on what banks are recommending. If they recommend fixed, I go variable. If they recommend variable, I go fixed.
    The reason: Banks are in it to make money so they will make recommendations on what seems to be a scare tactic in terms of “look where rates are at now.. you should lock in before they get any higher” while as I look at how rates have moved over the past decade or two and see that generally, rates don’t seem to go past a certain point much so if the rates are currently high, they should, over the 5 year period, go down.
    However, I can see how some people may not be able to tolerate the fluctuations in a variable rate if they don’t build up some room in their budget for the changes that might happen in the short term.

  2. I’m no mortgage specialist, but the variable option almost looks superior regardless of your situation.
    If we can trust this York U/Manulife study (, “88.6 per cent of the time during the past 50 years” floating rate option was ahead of the 5-year fixed.
    Plus, as I understand the variable option fixes your periodic payments to a set amount, and it’s the interest portion within the payment that will grow/shrink accordingly – this may compromise only the length of total amortization period (long-term), but would not affect immediate budget.
    Are there truly cases then where you could be 100% sure that fixed is a better choice? Or does it really come down only to one’s risk tolerance?

  3. Hi Patrick,
    I love the contrarian in you. :)
    You’re right in that rates are cyclical, just like the economy.
    I’m not sure if there’s enough data to validly test this, but theoretically it might be possible to “play” the rate cycles. (i.e. go variable when rates are “high” and go fixed when rates are “low.”) You’d obviously have to correctly define what “high” and “low” are.
    That said, this is clearly akin to timing the market. It doesn’t work for 99+% of people in the stock market, and it probably won’t work too well in the credit market–although I’ve never backtested it or seen any related studies.
    Cheers, Rob

  4. Hi Tele,
    Thanks for the comments. The research you mentioned is Moshe Milevsky’s widely quoted study from 2001. Based on this study, most people do assert that variables are the best option.
    Since the time period of his study, however, the spread between fixed and variable rates has narrowed significantly.
    Also, if your budget or GDS/TDS ratios are really tight–and you don’t have a “hold-the-payment” feature (many lenders don’t)–then variable is probably NOT the way to go.
    With respect to being 100% sure of a fixed. The answer is no. It’s impossible to be 100% sure of long term forecasts so we have to go by the percentages.

Your email address will not be published. Required fields are marked *

Copy link