After our 10-year mortgage story, one of our readers posed the question of how often 10-year mortgages were a better choice than 5-year mortgages.
We did a study on this a few years ago and decided to update it for this story.
To see how 10-year terms sized up, we:
- Looked at all available 5-year fixed rate data back to 1967–the earliest 5-year data readily available.*
- Added a 1.25% premium to 5-year fixed rates to arrive at an approximation of 10-year rates.
- Compared the 10-year rate to the average five-year rate a borrower would have received for two consecutive 5-year terms.
The data showed that homeowners would have come out ahead by choosing the 10-year term in only 49 out of 508 months.
While not a strictly scientific result, this statistic (1 out of 10), is similar to findings in the prominent fixed vs. variable studies. In other words, it appears borrowers don’t win long-term by paying big rate premiums for “safety.”
Some people will nonetheless say it’s different this time, and feel it’s worth paying 1.25% more for five extra years at a pre-determined interest rate.
There’s no denying that rates are at all-time lows today. Moreover, the odds of rates rising in the short to medium term appear greater than the odds of rates falling.
Nonetheless, the available data and modern low-inflation monetary policy both support the premise that 10-year mortgages are too expensive. Therefore, for most, it appears wise to avoid them.
Sidebar: Special thanks to the Bank of Canada for the data used in this study. * Data Notes: Prior to 1978 the conventional mortgage rate data was a simple average of rates charged by a number of large institutional lenders, including chartered banks for residential mortgage loans as at midmonth. Between 1967 and 1970 the average 5-year rate data also includes rates for 25-year mortgages.
From this statistics, it appears borrowers don’t win long-term by paying big rate premiums for “safety”. It appears wise to avoid 10-yr Mortgage..
Thanks for taking the time to do the study. I would have figured that through the 70’s and 80’s there would have been many more times that the 5/10 year difference would have been enough to make it more like a 30-50% chance that the dual 5 year terms would have come out more expensive than the 10 year term. Looks like that isn’t the case.
Were the 10% clustered in recoveries from recessions, as we expect to happen in 2010-11?
Good question Luke.
This just goes to show though that mortgage costs are a lot about risk. If you are willing to shoulder the risk of rate increases, you generally get a better deal.
Still though, other reasons a 10 year fixed might be good, is if you’re refinancing a rental property: that option gives very good cost control. Also, there are different ways to manage risk. (One of my favourite is to have more than one mortgage on a property …)
Thanks for doing the data crunching!
I did a quick and dirty comparison. I took the average yearly interest cost from the Bank of Canada PDF, which goes back to 1951 but has similar disclaimers about the data as presented above. I found that at a 0.5% premium for safety the 10 year worked out better about 30% of the time. At a 1% premium the 10 year works out only 14% of the time. Lastly at a 1.5% premium the 10 year only works out at around 6% of the time.
Just to check I tried 1.25% and ended up at 10.5% which is pretty consistent with above considering my math quality and just doing averages of the years.
To try to answer Luke, the majority of the success was during the run up of rates in the 70’s and 80’s, and even at the 0.5% premium, there have been none that I saw since about ’85.
what does it mean when someone say more than one mortgages on a property?
Traciatm: good idea, I was about to suggest that :) It’s interesting that even a small premium still results in a more expensive mortgage over half of the times. If the next 10 years are anything like the past it seems like it would take some very close rates to make the longer term cheaper.
I think the difference between today and our recent past is the government’s substantial and accelerating deficit spending. This will inevitably lead to significant inflation which could drive up rates like a rocket sled on rails, and at that point, the 10 year mortgage borrowers will be much better off. To me, the additional upside of taking a five year term represents a fraction of the potential downside of having to reset your interest rate in 5 years.
In these uncertain markets and with rates at all time lows, a small premium for an extra 5 years seems well worth the cost.
Could you please elaborate more on two mortgages for a single property. I have a current mortgage variable – 0.6 which I’ll port to a new property. Then I would need to take additional mortgage and I’m thinking fixed 5y. Does that make sense? Or convert all right away to 5y fixed?
David I bet the stimulus merely gets inflation back to normal. The BOC will not allow hyper inflation like the 80’s.
If inflation and rates rocket up it could be short lived. Maybe 2-3 years. After that the 5 year mortgage should again prove to be the cheaper long term option.
Remember, a 10 year mortgage must be cheaper for the full 10 years on average. That gives the 5 year plenty of time to beat the 10 years 1-1.5% higher rate.
Would your analysis be the same for commercial mortgages?
Mike & Brane:
More than one mortgage on a property is about getting a portion of your mortgage at a variable rate (higher risk, but perhaps lower cost), ideally with a readvanceable portion and another portion at a fixed rate. If interest rates move substantially, you can rebalance your payments so that you are paying off the more expensive option most aggresively. When I first did this, I got both from the same financial institution in the form of a fixed wrapped inside a readvanceable/LOC mortgage.
However, you can also get them from separate lenders. Caution though that in this case, there may be legal costs. Also, not all lenders want to be in “2nd position,” so your choice of lenders may be a bit more restricted. Finally, I got the impression that for all this to work, the total LTV had to be under 80% (and maybe even more in the current market?)
In any case, this doesn’t work so well if you’re maxed out on your LTV and amortization, ’cause then you have very limited options to change your payments when interest rates change.
My mortgage broker is the one who suggested this strategy: maybe Rob & Melanie can comment in a post?
Interesting study, but I’d like to see how much better the 10 year mortgage would have been during those 49 years. If during the other 459 years the difference was generally small, but the 49 years where the 10 was better had a larger difference, then it still might make sense.
Have there even been mortgages for 459 years Kenny? :)
haha, typing while tired. Replace those years with months :). I wonder what was considered a good interest rate 459 years ago?
When things were stable, it could be less than 5% (I don’t know about mortgages since those probably didn’t exist in the form we know).
Just bought an income property with my 24 yr old son, minimum down…5%, 3.39% for 5 years… Now I am thinking maybe we should have gone for 2 or 3 years and refinance in 2013-14 for 5 yrs or more and probably still good rates…Mortgage expires in Oct. 2016 at which time I suspect rates will have been on the rise already.
Can you update this article for 2012? BMO has a 5 year 2.99% rate now but 10 year rates are 3.78-3.99. So which is better now? In 5-10 years, do we believe the rates will still be this low?
I would like to know how folks see it today with such low 10yr rates as well.
Yes Jody I agree- knowing that the 10 year is at a historical low rate- even though it may stay this way for another one or two years- having the piece of mind knowing that I won’t have to worry about rates for another 10 years- sounds like a good time to lock in. It does seem however to go against the grain from what I read above. Does anyone think the 10 year will go down further as of March 2012?
Its a very specific oriented decision and there is no way to provide general commentary on something like this. I am moving from an excellent variable to a 10 year because with a large mortgage, three kids and lots of expenses I need some security. Everyone has an opinion on which way rates are going to go and honestly nobody knows. Have to go with your gut and look at your personal situation, sometimes we pay more for security. Also, the ten year buys you security if the housing market tanks a bit and renewing could be a problem for some if they haven’t paid down enough of the mortgage – this is something I think is being overlooked here.
Well the gap is smaller now. Talked to Mortgage Broker. Because we have a larger down payment 5 year is 3.09% (silly they penalized you for that, but it’s not ‘insured’). 10 year was 3.89% but broker said he’d do 3.84%. So with a 0.75% hike and only one way for rates to go we’re doing the 10 year term. How much money would we be talking about saving on a 5 year anyways? I didn’t see an amount actually saved. Is it worth possibly saving a few grand if the rates could go up to 6%? Rates are artificially held down and can only do so for so long and people have a really short term memory of days when rates have been much higher. Heck at one point it was almost like a credit card rate.
“only one way for rates to go”
Which way is that? Sideways?
Hi Ben, Thanks for the post. Here’s a related story that might answer some of your questions:
5 vs 10-year fixed
In the end…it is also about if the customer can sleep at night. If their risk tolerance is low, then the numbers don’t mean as much to them as the security.
Today the difference between the 5-year and 10-year in under 1%….would be great to see an update on this post with these rates.