If you had to pick one rate for the rest of your mortgage life, and your only two choices were a 5-year fixed or a 5-year variable, which would you take?

Most mortgage professionals (us included) would take the variable, citing long-term low inflation and historical rate studies, like those found here: Fixed or Variable…

But that doesn’t mean 5-year fixed rates are irrational.  On the contrary, longer-term fixed rates are quite suitable for borrowers who can’t withstand interest rate shocks.

Regardless, rarely a week goes by that we don’t see mortgage professionals paraphrasing studies that show variables are better 88-90% of the time. (In reality, 77% of the time is more accurate, assuming you negotiate good discounts. See this).

Those are good odds any way you slice it.

Yet, people routinely lose sight of the fact that 5-year fixed rates win out 10-23% of the time. That’s important to remember—especially in an environment like today, where prime rate has climbed while fixed rates have become increasingly competitive.

To see how the two rates stack up long-term, have a peek at the historical chart below. It shows simulated mortgage rates going back 30 years. The rates are simulated because today’s discounts are applied to past data.

The chart’s intention is to roughly approximate how fixed and variable rates might have performed over time with aggressive discounting. Moshe Milevsky, the father of Canadian mortgage research, performed his own simulated look-back here.

(Click to enlarge)In the above chart, the blue line represents the 5-year fixed rate while the white line represents a future moving average of the variable rate.

At any given point, the white line shows what variable rates would have averaged over the following five years. This enables a comparison of fixed and variable rates at specific points in time, using standardized rate discounts.

Naturally, discounts like 0.70% off prime and 1.70% off posted haven’t been around for 30 years. In addition, lender spreads and monetary policy have changed over time, which adds a challenge to normalizing rates (like we’ve done).

Nonetheless, “prime” and “posted” have long been the basis for variable and fixed-rate pricing. The only question is, what discount you choose to apply (it can be argued that past rates could have been significantly more discounted than they were).

At a minimum, the chart shows various times when highly-discounted fixed rates might have performed better than variable rates. That’s true even in cases where the fixed rate was notably higher than the variable rate to start. This seems to happen most often just as prime rate starts increasing from a trough in a rate cycle.

With respect to today’s rate environment, it just so happens that prime rate has started increasing from a trough in a rate cycle.

Long story short, variables are still a great bet for many. But, if you:

• Have a tight budget
• Are nervous that prime rate could exceed analyst rate estimates in 18-24 months
• Find an amazing deal on a 5-year fixed

…then no one can blame you for paying more for the insurance of a fixed rate.  You might get an earful from variable-rate proponents citing their probabilities, but as George Boole once said: “Probability is expectation founded upon partial knowledge.”