An oft-cited 2001 study by York University finance professor Moshe Milevsky reveals that from 1950 to 2000, a variable-rate mortgage would have beaten out a fixed-rate mortgage almost 90 per cent of the time.

But with interest rates at incredibly low levels, many commentators are expecting rising rates from here on in. Does that mean it’s time to lock in at a fixed rate?

If we look at the last 30 years, interest rates have gradually fallen. In August of 1981, the bank rate was 21.03 per cent. By the end of 2009, it was only 0.50 per cent. It seems intuitive that if rates are falling, a variable-rate mortgage gets cheaper and cheaper. Compare that with the sucker who locked in at a higher rate, right?

It turns out there’s more to it than that.

Prof. Milevsky’s research concludes that borrowers pay a “premium for predictability.” Lenders tack on higher interest in exchange for locking in at a set rate. It’s like paying for insurance to protect you if interest rates go up.

Even if you could time the short end of the yield curve, which drives variable rates, it would be tough to use this to your advantage, Prof. Milevsky’s work suggests.

Suppose you had the foresight to accurately predict interest rates. Let’s assume your mortgage is amortized over 15 years, or three five-year terms. Let’s also assume you can switch from a variable-rate mortgage to a fixed rate once during each term.

How do you suppose you, the perfect market timer, did versus someone who simply stuck with a floating rate for 15 years? The perfect market timer beat out someone with a fixed-rate mortgage 83.3 per cent of the time. But someone who stuck with a variable-rate mortgage for the entire 15 years did even better: They beat out the fixed-rate mortgage 88.1 per cent of the time.

So with a premium for predictability, and an uphill battle to capitalize on timing the interest-rate market, does the “all variable, all the time” strategy still trump all?

Well, who better to ask than Prof. Milevsky, who told me, “The recent numbers haven’t changed any of the main 2001 study conclusions around the benefits of floating over fixed, although borrowers should be aware that regardless of the mortgage they select, they are now paying abnormally low interest rates. So, if they are amortizing their payments over 15 to 25 years, they should be made aware of the fact that at some point they will be paying more – and they better make certain they can afford it.”

Prof. Milevsky concludes: “What happened last week is yet another example of how difficult it is to time these moves. A few months ago, many people were predicting a steady increase in rates from the Bank of Canada. People were told to not go variable because the rates were going up. Low and behold, last Tuesday, the BoC [Bank of Canada] paused. People that locked in a few months ago in panic and concern now look back and wonder whether they overreacted. Bottom line: Don’t try to outguess the bond market or the BoC.”

While a variable-rate mortgage has the highest probability of outperforming a fixed-rate mortgage, the sensitivity to the borrower’s cash flow is of utmost importance. That’s why Prof. Milevsky calls it the “premium for predictability.”