Wednesday, September 9, 2015

Should You Take a Variable Rate Mortgage?

A fellow financial blogger asked my opinion about his upcoming mortgage renewal. He faces the same choice as many of us do: should you take a fixed-rate mortgage or go for the lower variable rate? The risk with the variable rate mortgage is that rates might rise. The answer requires surprisingly little math.

If rates stay the same for 5 years, then the lower variable rate will save you money compared to a 5-year fixed-rate mortgage. If rates go down, you’re even further ahead. Averaged over all possibilities, the average outcome is that you save some interest on a variable-rate mortgage. The worry, though, is the possibility that rates go up.

You can’t fully protect yourself against rising rates even with a 5-year fixed rate, because you’ll have to renew at a new interest rate after 5 years. But you might hope to get your balance down enough that you could absorb an interest rate increase in 5 years.

The real test of what you should do comes with looking at a terrible outcome. Suppose you could get a 2% variable rate today. What would your payment be if the variable rate shot up to 7%? This isn’t a prediction. We’re just looking at what happens in a scary scenario. It’s your reaction to this scenario that should drive your decision.

After calculating your mortgage payment at 7% interest, did you throw up? Did you look at it and know for certain you’d lose the house and everything else you own? Then a variable-rate mortgage isn’t for you.

Instead, did you look at the 7% payment and think that it would be no fun to pay this higher amount, but that you’d be okay? Then you can go ahead with a variable-rate mortgage.

A curious thing about this piece of advice is that I don’t know of anyone who has followed it. Some say it makes sense to them, but they never actually calculate the payment based on 7% interest. They just imagine a higher payment and declare whether they can handle it. This is useless without actually calculating the payment based on the much higher rate.

One explanation for not following this advice is laziness. Another possibility is pain avoidance. It’s no fun to imagine a much higher payment, and not calculating the higher payment avoids the pain. Whatever the explanation, deciding if you can handle a higher payment without actually calculating the higher payment amount is not helpful.

1. My personal opinion is that if rates shot up to 7% and the payment would make you throw up or you believe you'd lose your house, you probably shouldn't have the house you have. Sure, a fixed rate now might make keep you solvent for 5 years, but as Michael suggests, when you have to renew at 7%, it's unlikely you'd have built enough equity to avoid a punishing payment. Better sell the house now and downsize or rent!

Fixed rates feel like insurance to me. Odds are they will cost more, but it protects your from a rise in interest rates. If a big jump in interest rates would cripple you financially, you may need more insurance, such as a 10-year fixed rate. This would cost even more, but it provides more protection against rate increases.

1. @Returns Reaper: Good points. The 7% test is a good one to apply before you even buy a house. But if you've already bought the house,selling now may be an overreaction -- it depends how close to the line you really are.

I agree that fixed rates are a form of insurance, and the longer the term you go for, the higher the premium.

2. Rates are not going to 7%. What people seems to forget is we are at say 2-2.5% in rates. Let's use 2.5% for my example. An increase to 3% isnt just a 0.50% increase, in absolute terms it is, but in relative terms it is a 20% increase. An increase to 3.5% is a 40% increase in interest costs. So if you think rates are going to shoot up, you're mistaken. Rates will stay low for years until we see wage expansion in the private sector, which certainly isnt happening.

3. @Unknown: I don't believe you or anyone else has useful insight into the future of interest rates. Looking at interest rates in relative terms is not useful. If rates went down to 0.01%, you're saying that it would take a massive event to bring them up by a factor of 10 to 0.1%. This is nonsense.

4. I would say it becomes more and more difficult to raise rates the lower we get....IE Japan, and the US, and the number of other counties that are struggling to raise rates.

2. My variable mortgage is 1.9%, up for renewal in early 2016. I'll be going variable again.

Guess the first thing to do would calculate the probability of 1) rates actually going to 7%+ and 2) staying there for extended periods.

I doubt either will happen within the next 5-10 years.

(The long-term average of Canadian mortgages is ~10% and spent ~20 consecutive years above 7%.)

1. @SST: I doubt that my house will burn down, but I still have fire insurance. It's not about whether a steep rise in interest rates is likely. It's about whether you'd be OK in a low-probability bad scenario.

2. Structures have burned down for as long as humans have lived within combustible boxes, and the trend will continue. Insurance would be mostly for the accidental appearance of fire.

Mortgage rate hikes, on the other hand, have never been accidental and have always been consiously applied.

If the probability of a short window 5% rate hike is far too small then I won't even bother figuring out if I can or cannot afford it.

Neither do I fret about what I'd do if I got hit by lightening, crash in a plane, or win the lotto.

Besides, isn't it law that you must have house insurance?

3. @SST: Interest rates moves are forced by conditions. The discretion of those who make the choice to move rates is very limited.

I don't think 5 years is a short window. Interest rates have certainly moved a great deal during 5-year windows in the past.

3. I believe that past history shows variable rate mortgage holders end up paying less in interest, so mathematically it's the better choice. But as you quite rightly point out, there's a huge downside risk.

SST seems to think it's 'unlikely'. But unlikely doesn't mean 'can't happen'. It happened in the 80's in Canada when rates went to 20%, and it happened in the past decade in the U.S. with balloon mortgage payments. In both cases, lots of people lost their homes.

Because I'm unwilling to lose my home to the bank, I've always locked in long term, typically for 5 years. And the stress test I do is to take my mortage balance at the end of 5 years and run it through a mortgage calculator at 15%. If I can handle the mortgage payments at that level, then I'm OK to proceed - because I know in 5 years at mortgage renewal that I'm not going to lose the house if interest rates get stupid.

1. @Glenn: Looks like you use a bigger margin of safety than I suggest. Maybe yours is the safer approach.

4. I guess I'm getting old. My first mortgage was a 5 year term at 7% and that was the lowest interest rates had been for decades. 5 year mortgages spiked briefly up to 9% in that 5 year term but averaged less than 7% overall. Subsequently I went for 1 year terms with generous pre-payment options. I always felt that plowing the interest savings (and extra cash) into reducing the balance was a good form of insurance against rising rates. Michael, I think one of the assumptions of your post is that the mortgage will take 25 or more years to pay off. I think the calculations change a bit when there are only 10 or less years left.

1. @Greg: Things change when the amortization period is shorter, but I believe my approach accounts for this. With shorter amortizations periods, payments are less sensitive to interest rates. But it still makes sense to check the payment in the even that interest rates rise substantially.

5. Rates also matter less with a shorter amortization on a mortgage.

6. Love your site - some great articles and in-depth analysis you can't get anywhere else.

Of course I agree that the first consideration when deciding fixed or variable is whether you can afford a steep rise in rates. But I was a bit surprised that you found that to be the end of the analysis.

The spread between 5yr fixed and variable these days is about 0.5%, which means the banks must consider the expected rate increase to be less than that amount over the next 5 years (since they are anticipating making money on the insurance portion of fixed rates). Of course, the banks might lose that bet and rates could go up much more than 0.5%, and if that happens there's no question that the banks could "afford" to stomach those losses but somehow I think the banks will find a way to make a profit even in that scenario. What will they do? Buy forward contracts to hedge the interest rate risk?

Is it possible for an individual to hedge their own variable rate mortgage with forward contracts, thus decreasing the spread between fixed and variable from 0.5% to say 0.2%? If you could effectively "self-insure" your variable rate mortgage for less than the banks are offering, would that tip the scales towards taking a fixed rate?

Thanks again for all your work maintaining this blog. It is a hidden gem of Canadian personal finance.

1. @Dan D: It's certainly possible to self-insure your variable-rate mortgage against interest rate increases using derivatives based on interest rates, but I don't know if an individual can do it cheaper than the premium a bank offers. Derivatives are a world where even experts get burned. The vast majority of people should stick to conventional mortgages of one type or another.

7. One could always take hedge actions against rising rates in order to keep their payments similar.

A true life example would be a relative who, during the '80's ramp up, went from bank to bank borrowing and depositing upon higher and higher rates. She made \$100,000 in four years.

It can be done, simply and easily, if you view increasing rates as an opportunity not merely a personal detriment.

1. @SST: You'll have to explain that one. You portray bankers as idiots who lend at lower rates than they pay. This hasn't been my experience.

8. Bankers are a slick bunch, of course they will take whatever means possible to come out on top. But in an environment of rapidly increasing rates, I can only assume there was an available opportunity. I am unaware of the exact mechanics she employed but will definitely inquire.

Probably similar to how a house flipper can make \$100,000 in a rapidly declining rate environment by simply sitting and waiting. (I used to live in Alberta, I got to witness this phenomena first hand.)

Or perhaps it was like when he price of silver collapsed to sub-\$5 levels and people were going to banks to cash in their Montreal Olympics \$5 silver coins for "free" money.

The take away is, there is more than one way to skin a cat, and if you view rising interest rates (or any market condition) as a one-way street (i.e. from your wallet to the bank), then you probably aren't going to be having a good time.

1. @SST: I think I'll stick to strategies that don't require me to outsmart banks.

9. Maybe it's because we have an optimistic bias. I always wonder how much my portfolio will rise next year, not wonder IF it will rise. Surely I'm not alone? Here's a commercial I heard about on the Freakonomics blog that tackles this bias: https://www.youtube.com/watch?v=o3pFHPgH3oU

1. @Gene: The optimistic bias is a good reason to think that many people won't even think about the possibility of mortgage rates rising. However, once it is pointed out as a possibility, and the idea of choosing a much higher rate as a test, you'd hope that more people would check on the payment for the higher rate. Maybe part of the optimistic bias is an outright refusal to think about bad outcomes, even once they enter our minds.

10. The great thing about 7% mortgage rates would be house prices would decline, likely by a lot, although payments could stay relatively the same.

As well, the quality of housing stock would also rise; 7% rates might just be the death knell for all those million dollar crack shacks in Vancouver.

Much higher rates could also stem the rise in fraudulent practices in the mortgage industry (e.g. Home Capital).

No one wants to pay more, that's just human nature, but higher rates -- fixed or variable -- could usher in a higher degree of stability.

1. @SST: These would all be positive developments for those who don't yet own a house or who don't have a mortgage. For those up to their eyeballs in mortgage debt, much higher rates would be a disaster.

11. Do you think it's worth comparing the current interest rate to the long term average when considering variable vs. fixed?

1. @John Ryan: That's one data point, but I don't think this comparison leads to a final decision. The main thing is to decide what is the highest interest rate you want to protect yourself against and check out the associated payment. I see people talk about the payment if rates go up half a point. But does that mean you don't mind losing your house if they go up a full point? People really need to think about the worst downside they consider too likely to ignore.