A Detailed Look At Canada’s Mortgage Stress-Test Rate

As of January 1, 2018, everyone who borrows from a federally regulated mortgage lender must qualify using the Mortgage Qualifying Rate (MQR), except for renewing borrowers (who aren’t subject to this new standard if they stay with their existing lender).

The MQR, which is also commonly referred to as the ‘stress-test rate’, was designed to ensure that anyone taking on mortgage debt at today’s ultra-low rates can afford for them to rise in future (and using today’s MQR, that means all the way to 4.99%).

While this is a sound and prudent approach in theory, the method that is used to set the MQR is seriously flawed. And given the impact that this all-encompassing benchmark rate will soon have on the mortgage market, it should be modified ASAP.

In today’s post we’ll start by taking a look at how the MQR is currently set. Then we’ll use an example to contrast the MQR’s impact on a person’s ability to qualify today with the impact that actually borrowing at the MQR rate would have on that same person five years hence. Put simply, we’ll conduct a stress test for the stress-test rate.

Why the Current MQR Methodology Is Not In A Borrower’s Best Interest (pun intended)

When our regulators came up with the MQR they simply took an average of the Big Six’s posted five-year fixed rates. But simple is not always best, and this was a flawed approach.

For starters, the Big Six’s posted five-year fixed rates aren’t actually used for lending, except maybe to renewing borrowers who fail their Bank’s laziness test. Instead, these posted rates are primarily used to inflate the size of Big Six fixed-rate mortgage penalties.

Given its importance, the MQR should be based on real market rates that fluctuate in response to real market conditions. This change would increase transparency and allow market watchers to better forecast the MQR’s future movements (unlike today, when Big-Six posted rates move to the beat of their own drum).

If the objective of the MQR is to ensure that borrowers are able to handle real increases in market rates, it should be obvious that any test of that ability should be based on those same underlying rates. The current MQR formula fails that test.

My other issue with the current methodology used to set the MQR comes down to fairness.

Basing the MQR on the Big Six’s posted five-year fixed rates gives this sub-group of lenders de facto control over consumer access to mainstream residential mortgage borrowing, and that creates the potential for conflicts of interest (pun also intended … and yes, I’ll stop those now).

This flaw could easily be addressed by basing the MQR on rates that are used for actual lending, with a premium added that meets the regulator’s objective. This simple change would transfer control over the MQR from the Big Six banks to a combination of the regulator and the market instead.

More broadly, we should remember that any policy that favours the Big Six hurts all borrowers, regardless of which lender they ultimately borrow from. Mortgage rates would be higher and borrowing terms and conditions much more onerous without stiff lender competition to keep the Big Six on its toes. While our regulators may have had no choice but to make policy changes that unintentionally favoured the Big Six in order to slow the market, that excuse doesn’t apply to the MQR calculation.

Bluntly put, the MQR calculation should be modified so that it is more transparent and fairer for all market competitors (and before January 1).

How Much Stress Should the MQR Test?

Today’s MQR requires borrowers to qualify at a rate of 4.99% but is that the right rate? Put another way, how much stress should the MQR test?

Let’s use an example to help answer this question.

Assume that Jeremy, who is otherwise debt free, is buying a $500,000 house and is applying for a $300,000 mortgage with a five-year fixed-rate of 3.00% amortized over 30 years. That mortgage will cost him $1,262/month.

After January 1, Jeremy will be qualified at the MQR rate of 4.99% amortized over 30 years. This stress tests his ability to pay $1,599/month, which is 27% higher than his initial payment today.

Next, let’s fast forward five years and assume that mortgage rates have risen to that point where Jeremy is concerned about being able to afford higher payments.

After five years Jeremy will have reduced his mortgage balance to $266,628. As such, if his mortgage rate at renewal is 4.99%, his new payment (based on his remaining 25-year amortization period) won’t be the $1,599/month that was used for his original stress test. Instead, it will be $1,549, which is only 23% higher than his original payment.

Furthermore, if Jeremy is having trouble managing a bigger mortgage payment, he also has the option to re-extend his amortization back out to 30 years. If he does this, his new mortgage payment will drop to $1,421, which is only 13% higher than his original payment. While re-extending his amortization isn’t ideal, if Jeremy is under real financial stress, this age-old tactic can be used to help mitigate his strain.

The original stress test required that Jeremy be able to afford a payment-shock increase of 27% at renewal, but as we now see, his renewal rate would actually have to be higher than 4.99% before that level of payment increase actually occurs. Using the numbers outlined above, on a straight renewal with no change in his remaining 25-year amortization, five-year fixed rates would have to rise to 5.32% to trigger a 27% increase in Jeremy’s payment. And if he re-extends his amortization period to 30 years, Jeremy’s rate would have to rise to 6.08% before his monthly payment would increase by 27%.

This is an important point because while it appears that the stress test is designed to ensure borrowers can afford for rates to rise to 4.99% at first glance, it is the mortgage payment, not the rate, that impacts Jeremy’s affordability at renewal. In that context, mortgage payments based on an MQR of 4.99% today are actually equivalent to renewal rates in the 5.32% to 6.06% range five years hence.

That brings me back to the key MQR question: how much stress should the MQR test?

Is it reasonable to assume that interest rates will increase from 3% today to between 5% and 6% at renewal? Especially without factoring in any corresponding increases in Jeremy’s earnings over that five year period?

Regular readers of this blog won’t be surprised to read that I think mortgage rate forecasts in the 5% to 6% range in five years’ time are aggressive. But if our regulators won’t explain how and why they arrived at an MQR of 4.99%, how can we properly evaluate the reasonability of their assumptions? Shouldn’t the stress test also be subject to a smell test?

As we have seen with the recent small business tax proposals, initial positions can, and often do change.  Let us hope for some change in this case as well.

The Bottom Line: When the MQR bites the market harder in the new year, the methodology used to set it will be more heavily scrutinized, and will not be easy to defend in its current form. Our regulators owe it to Canadians to make the MQR a fairer benchmark that is based on real market rates instead of arbitrary posted ones, and to make the rationale behind it more transparent. Over to you Jeremy.