How People Choose Between Fixed & Variable Rates

As much as academics advise against it, people (consciously or subconsciously) try to predict interest rates before choosing a mortgage.

But, interestingly, they don’t look very far into the future when making these forecasts.

“…Households are forward-looking over relatively short periods of time,” research shows.

A 2015 international study by Cristian Badarinza, John Y. Campbell and Tarun Ramadorai found that “the spread between the [fixed rate] and the average rationally expected [variable] rate over the next year is often a better predictor of [variable-rate market] share than is the spread between the [fixed rate] and the current [variable rate].”

In English, that means most people do two things:

1. They create their own rough estimate of where rates are going in the next year or so
   • If they think variable rates will fall, they more often choose variable, and vice versa
   • This is usually a mistake as research shows that mortgagors can’t divine rate direction with any reliability
2. They evaluate the difference (a.k.a., “spread”) between fixed and variable rates
   • If that difference is too small, people are less willing to take their chances with a variable rate

The research shows that folks weigh #1 more heavily than #2 when choosing a mortgage.

Are People Thinking Straight?

“…Households do not seem to anticipate longer-term rate movements in the manner that would be required to minimize the lifetime costs of their mortgages,” say the study authors. In other words, peoples’ rate predictions aren’t saving them much money.

They also find that long-term rate expectations don’t sway borrowers’ decision-making as much as short-term rate expectations. Perhaps that’s because people feel they have less chance of being right about rates far out in the future. In this case, “…the current rate spread is [a more] important influence on mortgage choice.”

That’s essentially what’s happening right now. Despite market indications that rates should remain mostly flat to lower in the next few years, prospective borrowers are fixated on today’s razor-thin spread between fixed and variable rates.

This is reflected in the above-average number of people locking in. It’s also showing up in the number of people searching Google for the best fixed rates (as opposed to the best variable rates). In 2019, that number has been three times greater than what it was in the same time period of 2018.

Historically, the spread between fixed and variable rates is “normally” at least 75-125 basis points. When the spread is that wide, the authors find that two things tend to happen:

1. “…Borrowing-constrained households seek low rates [which allow for lower payments] in order to maintain the level of current consumption…”
2. Borrowers choose variable rates “to increase the size of the house they can buy when constrained by bank limitations on mortgage [debt-service] ratios.”
   • Note: Despite Canada’s federal “stress test,” it is still possible to qualify for a bigger conventional mortgage if you choose a variable rate. This happens more often when:
     • the best fixed rates plus 200 basis points are higher than the 5-year posted benchmark rate, and
     • the best variable rates plus 200 basis points are less than or equal to the 5-year posted benchmark rate.
   • This variable-rate qualifying “advantage” is more pronounced when the fixed-variable spread is large.

That said, both of these points are largely moot today (for prime borrowers). When the fixed-variable spread is as narrow as it is currently, many borrowers simply figure, “What’s the point?” and just go fixed.

(Sidebar: Needless to say, these aren’t the only decision factors affecting borrowers’ term selections. Other things matter too, like prepayment penalty differences between fixed and variable mortgages.)

Economists Sway People

The econo-experts you see quoted in newspaper columns and business TV seem to have a real influence on people’s mortgage decisions. “The [variable-rate market] share tends to increase when professional forecasters expect short-term interest rates to decrease during the next year,” say the researchers.

The fact that long-term economic forecasts are semi-worthless is beside the point. People place weight on them.

Regardless, “…the current [fixed-variable] spread remains statistically significant, even in the presence of this effect.” Hence, we can be fairly certain that until the yield curve becomes more normal (i.e., until long-term rates are once again higher than shorter-term rates), demand for fixed mortgages will exceed historical norms.