Blessed by the Potato

The rules of curling come with an expiry date. I guess it’s to manage the cadence of rule changes, or maybe to force people to actually check for rules updates every few years, but whatever the reason, they expire. Specifically, they expire at the end of this month:

The word is that Curling Canada is going to fold in with the World Curling Federation and stop having a separate rule book, starting with the next iteration of the rules coming out for 2026/2027. That is expected to be released the week of September 7th, according to a post from CurlBC.

This means there’s going to be no rules for the first week of September, 2026!! For most players and clubs, that’s fine — our ice won’t even start going in until the end of September, and our season doesn’t start until after Thanksgiving in October. However, there is a summer league happening in Oakville, which I am a part of.

The Purge: Curling Edition is on! (or will be in a few weeks)

So, does anyone have a battery-powered floor polisher I can borrow for September 1st? :)

The RRSP is the most misunderstood tax shelter in Canada. It holds pre-tax money, which is not intuitive to think about at all. So when it comes to the Home Buyer’s Plan (HBP), there’s similarly a lot of conflicting articles out there that don’t quite analyze the options right.

First off, what is it? The Home Buyer’s Plan is a relief valve for RRSP investments, and an incentive for first-time home buyers: a first-time buyer can take up to $60,000 out of their RRSP when buying a house (presumably for the down payment but the funds can be used for anything — moving costs, furniture, renovations, just sticking in your savings account, the key test is that you bought a home, not specifically which dollar did which job). That relief valve means that people who aren’t sure if they should be saving for retirement or for a first house can utilize their RRSP for saving/investing, and still get a decent chunk of the money back out later if they end up buying a place. Then if you do take the money out of the RRSP, you have to put it back in over the course of 15 years, with a little delay (2 years) before you have to start. If you fail to make the repayments, the amount gets added to your income and you lose the RRSP room, just as though you had made an RRSP withdrawal in the year you skipped the repayment.

The FHSA is hands-down better in every way, but has limited contribution room, so a new buyer will likely have both a FHSA as well as some RRSP contributions made. And fortunately, you can use both.

Now if you’re cutting it close on having the funds for a downpayment, the HBP is a nice little incentive: because it’s pre-tax money, with a bit of planning ahead (you have to have had the funds in your RRSP for at least 90 days), you can take your downpayment savings, throw it in an RRSP, get your tax refund, then use both the original after-tax amount and the refund for a downpayment.

What if you don’t need that tax refund/pre-tax aspect to get more scraped together for your downpayment? That’s where it gets tricky: you’ll find articles describing the HBP as “an interest-free loan to yourself” (ignoring the opportunity cost of investing in your RRSP), or “like earning your mortgage rate in your RRSP”, which actually sounds worse than leaving it in and investing in equities. You may also see some comments saying if you don’t need it, don’t bother: if you’re going to have investments left over, wouldn’t you rather have investments in your tax shelter than out of it? And it’s never clear if these are the correct framings, or if someone is forgetting about the pre-tax nature of RRSPs, or just saying things to fill the silence.

So, let’s set up the analysis: you’ve been renting and investing the difference for over a decade, so have plenty saved up to put down on the house. If you don’t need that pre-tax boost to meet your desired down payment, does it make sense to use the HBP anyway? You need to figure out where to source the last $60,000 of downpayment (or land transfer tax, or pre-move-in-renos, whatever): from your non-registered, TFSA, or RRSP? Let’s put $60,000 (nominal) in each, and see how things work out for a simple model.

From our asset location breakthrough we should recall: the RRSP holds pre-tax money, so $60,000 coming from there is equivalent to less being taken from a non-registered or TFSA account. We should see this fall out when we calculate out the impacts. So this should work out best by taking advantage of the HBP. But that HBP also comes with a pesky repayment requirement and up-front paperwork and sitting on hold with your brokerage. Will it be worth the hassle and brainsweat?

Here are our three scenarios: take from the non-registered account; take from the TFSA; take from the RRSP-HBP. In each case, we’ll have 1/15th of the withdrawn amount ($4000) each year as cashflow to put into the account that was used (ignoring the grace period). In our tax shelters, our investments will increase at 6%. In our non-registered account, the tax rate will be 15% (i.e., a 30% tax bracket but with investments taxed at half the rate), for a 5.1% after-tax growth rate. In our RRSP, at the end of 15 years we’ll withdraw it and pay tax at 30% to get everything in after-tax dollars. I put everything in a Google sheet so you can follow along (no it’s not publicly editable — download a copy for yourself if you want to play with your own numbers).

1: Use the RRSP HBP

We pull the $60,000 out of the RRSP. The $60k in each of the non-registered and TFSAs keep growing. After 15 years, we have: The TFSA amount grew to $143,793. The non-reg, thanks to tax drag, grew to a just $126,530. The RRSP growth and slow rebuild over 15 years led to $93,104 pre-tax, but melting that down to make everything in after-tax dollars leaves us with $65,173. In total, the net worth at the end was $335,496.

2: Use the non-registered account

We use $60,000 of non-registered funds, leaving our tax shelters alone. The RRSP and TFSA each grow to $143,793 nominal (as the TFSA did before, good sanity check), but after melting down the RRSP was $100,655 after tax. Meanwhile the non-registered grew to $86,967. That leaves the total net worth for this case at $331,416.

3: Use the TFSA

Come on, you knew this was going to be the worst option — that logic of “if you have investments left over, wouldn’t you want them to be sheltered” is much more straightforward here. The re-built TFSA is worth just $93,104 at the end, with the other accounts that’s a final net worth of just $320,289.

Conclusion:

The HBP is useful for people who don’t have enough funds available for their downpayment: they can use pre-tax money for a little boost. Even if you do have enough, using pre-tax money and letting your after-tax non-registered investments grow does have an advantage: with fairly reasonable assumptions, about $4080 of future (i.e. 16 years from now) value — at that same rate of return, about $1850 of present day value. Now that is nothing to sneeze at, but for someone in the harried stress of buying a house, is that worth the headache of all the HBP paperwork and then repayment schedule? (Actually, yeah, it’s not that much work and not often you can get almost two grand just from tax optimization).

The (future) value of the HBP changes a bit as the assumed rate of return and tax rate change.
Rate of return 4%: $2885 and Tax rate 50%: $5101
Rate of return 5%: $3525 and Tax rate 50%: $6392
Rate of return 6%: $4080 and Tax rate 50%: $7636
Rate of return 7%: $4511 and Tax rate 50%: $8796
Rate of return 8%: $4774 and Tax rate 50%: $9829

But it’s always enough that the headache is likely worth it.

Not modelled: what happens if your tax rate changes over time.

Bonus:

What if you don’t invest in equities, and instead your non-registered is in bonds/GICs at your full tax rate? Then the HBP actually works against you: the tax drag on your non-registered account is too much for the pre-tax bonus of pulling the RRSP funds out to overcome, and you’re better off not using the HBP. So if you’re not going to invest the money (or use it on the house), don’t do it!

In 2011 I refined my ranting about high house prices and released the rent-vs-buy calculator, helping people to model the complex moving parts of comparing owning to renting: both have unrecoverable costs, both have cashflow needs, both have ways of building equity. A few key innovations were that it allowed someone to model what might happen if interest rates re-normalized, and it explicitly estimated the cashflow and net worth differences so people could see the counterfactual: how much equity a renter could build up in the stock market by investing instead of buying a house.

Moreover, at the end of 2011 I faced the same decision point myself: after finishing grad school, I was moving back to Toronto, with a baby on the way. Were we going to hold our noses, buy an over-priced, under-sized place… or rent? In addition to the advantageous financial projections, the choice to rent had a few other key pros. One is that renting being cheaper meant we could afford a larger home: instead of trying to make a cramped condo or townhouse work, or driving ’til we qualified (which would have taken us right back to the L-dot), we managed to get a lovely 3-bedroom detached house walking distance from the subway (well, Line 4, but a subway nonetheless). It would also make the path to upgrading a bit easier if we had a second kid: we could focus on the house size we needed now rather than try to jump straight to a “forever home”. Though we put in a lot of analysis, it was all a bit moot: at the time of the move, I didn’t actually have a grown-up job yet. I was still a post-doc, making subsistence wages, hoping to find a job after landing in Toronto; Wayfare was self-employed, with all the complications that brings to getting a mortgage. So we never really had the option of buying a house, as we would have never qualified to borrow the insane amount of money one cost at the time.

So we rented, had Blueberry, and had relatively secure tenancy: it was 11 years before the dreaded notice came. But the notice to vacate came nonetheless, and one of the risks of renting reared its ugly head: we were forced to move. The timing was bad but not the worst: it was the summer between grade 5 and 6 for Blueberry, so she was moving schools anyway. But it was the summer: we only had a couple of weeks to find and secure a new place to get her registered in her new school, or risk having to up-end her partway through the school year. It was a sucky, stressful experience, but we did secure a new place on the last day before registration.

Over those 11 years, Toronto real estate went from pricey through ludicrous right into plaid¹. Back in the early days of the debate, I used to say how crazy the bulls’ projections were: 6 or 7% appreciation per year was going to make it so the average house was over a million dollars in short order. A million! For a regular house you still had to commute an hour to downtown from! That was so much money to spend on a house. And it actually came to pass and people just normalized it and accepted it (and many even cheered it), and even in hindsight it seems crazy to me that this is the timeline we’re on.

Anyway, two years before we got notice to move, I did a ten-year check-in: Rent vs Buy: So How’d That Work Out for You? where we found out that after a decade, someone who chose to buy in Toronto came out ahead of someone who chose to rent. It was a lot closer than people who don’t do the math think: when a house goes from $750k to $1.4M, nearly doubling in a decade, it’s pretty easy to mistakenly think the homeowner was ahead by almost $650k (or more as many a person who has never mathed it out will be quick to chime in, “’cause the renter paid all that rent with “nothing to show for it””). But renting was cheaper from day 1, so there was cashflow for the renter to invest, alongside the downpayment. Rents increased in the city, accelerating around that time (and going ballistic in 2022 and 2023, then settling back down lately), but not actually that far off the predictions we started from (though condos and anything that could be converted to hotel rooms or rooming houses experiencing much worse rent inflation for a while, now outright deflation in the correction). A key point though is the stock market also had a hell of a decade with all that ZIRP money floating around. So while the buyer did come out ahead, at what in hindsight was just about the peak of the market, it was “only” by about half the headline house price appreciation: a hair over $350k.

So 2023 comes around, and we have to make the decision again thanks to that dreaded N12. Rents had a pop in 2022/2023, but house prices are just cresting the top, while interest rates are high. Again, buying a house is not in the cards — with the rates (nearly 6% in 2023!) and the stress test, we’d never qualify, even with hefty down payments from all that saving-and-investing the difference. The math says renting is expected to do better, and we want a detached house again regardless, so renting is the easy choice.

Here we are in 2026, and after just 3 years we are once again getting the heave-ho from the landlord: risk acknowledged a priori still sucks when it’s realized post hoc. But it’s a good time to check in: how’d that rent-vs-buy decision work out in the end?

Ben Felix has a video out called The Reckoning where he looks at the results using aggregate data across a number of metros up to 2025. But how did it actually work out specifically for a Toronto family-sized detached house?

Terrific, as it turns out. While things looked a little depressing in 2021, it has been all good news for the renting crowd since then. The stock market continued to go up: a basket of index funds is up roughly 80% over that time. House prices have come back down a hair: the $1.4M house at the peak is now about $1.3M. Aggregate stats for the city show more of a decline, but shoebox condos have been hit harder than suburban detached houses, and places like Brampton hit harder than North York or Markham, so for us in the north-east end of the GTA, it’s more of a slight decline/flatline than an outright market crash… this pocket also didn’t accelerate as badly in 2021/2022 in the first place.

But as I’ve said all along, The Big Misunderstanding^TM is that prices had to crash for renters to come out ahead. They don’t. They just had to stop going up like crazy. Indeed, right from the very start the rent-vs-buy calculator said if you used sane assumptions for house price appreciation, renting would come out ahead. If you tweaked the numbers to see what buyers were implicitly assuming to make them choose to buy over renting, they were baking in roughly 6%/yr appreciation — forever. It was astounding that they got 6%/yr for so long (and more in some years!), but now a few years of zeros following that growth has brought the average long-term growth to closer to 3.5%/yr.

Over the 5 years since 2021, the owner paid off a bit more of the mortgage to build some equity… but lost $100k to the slight decline in market value. Meanwhile, the renter had more cashflow to work with because the owner was shelling out more for interest. Oh, and that spike of . And insurance. The big thing helping the renter though is the stock market being up roughly 80%. The stock portfolio alone made something like $550k² for the renter, dwarfing the cashflow differences. The renter has rocketed from behind, and can now buy that now-$1.3M house with just a 5-figure mortgage³ if they so choose, while the owner is still looking at a mortgage balance of $260k, and another $65k or so of transaction costs if they also have to move.

To compare to Ben’s broader results, he presents it as a “wealth ratio”, which for Toronto was 1.37 in favour of the renter. That understates things: if the house price rocketed to $1.3M, with $1M of equity, that would imply the renter is $370k ahead, with a net worth of $1.37M (so our idiosyncratic results were worse, a wealth ratio of “just” 1.1, but house prices in North York/Markham have held up much better than those in the GTA aggregate). I think Ben really under-sold how impactful his results were with that table — those little decimals hide some pretty large effects and differences, especially with such high house prices. He’s also using a different period (25 years vs 15), though prices largely weren’t that expensive in 2005 so it’s amazing to see renters come out ahead at all in his example.

Anyway, there were of course some non-financial trade-offs. On the pro side, we didn’t have to worry about major house maintenance. When the ice storm hit and the pipes burst after the 7-day power outage, it was the landlord’s duty to fix it. I never had to call up a bunch of professionals, try to get a quote, then send 30 separate interac e-transfers for the job (which I’ve had to do to help my mom…). For a long part of our tenure, we didn’t even have to mow the lawn. Our cashflow needs were lower, so when Wayfare got sick we could still afford the house on one income (and weren’t facing a forced move/downsize on top of recovering from a rare disease). Those lower cashflow requirements also offered us the option of spending some of that surplus by getting a nice, fully detached house in a nice neighbourhood, which we never could have afforded if we bought. On the con side, there’s security of tenure: moving sucks. Moving because someone else is making you double sucks. Moving on a tight timeline because someone else is making you less than a month before school starts sucks triple. {scott_pilgrim_this_sucks.gif}

Framing it as a compensated risk makes it a little easier to handle. As much as moving sucked, and as much as I’m hating the search for a new place (esp. constrained to the same school district) and as much as I’m dreading the next move… we’re financially better off to the tune of $100k just for choosing to rent. Back then, $100k was considered a lot of money (now it’s just the extra a realtor will try to convince a buyer to throw at a place in a bidding war, barely a rounding error on today’s house prices). Outside the context of house prices, it’s still a lot of money. Moving was weeks of effort: nearly every moment outside of work was spent packing, tidying, shuffling things between houses, cleaning, unpacking, updating mailing addresses, updating bill payments… But for $100k ($50k/move), that’s the best-paid job I’ve ever had.

So to sum up, renting and investing the difference did leave a person in Toronto better off than holding their nose and buying. Prices are still way above where they started in 2011, but didn’t keep going at that white-hot 6-7%/yr rate, and with a bit of a correction ended up averaging out to just 3.5%/yr — only a hair higher than the long-term average assumption we started with. The stock market did amazeballs, and rents continued to take less cashflow than owning. As much as the big shift from being behind in the check-in 5 years ago to being ahead now was surprising, I’m really surprised this real estate market is still going: a tiny little correction, and no reckoning yet for the cap rates. After how far ahead the owners were in 2021, I’m shocked the scales flipped with just a soft landing. Zero blood in the streets and practically no industry bankruptcies.

Of course as nice as the vindication is, it’s highly bittersweet when sitting here with another N12 in our hands.

Still, let’s end the post on a high note: winner winner, chicken dinner! All the math and logic did eventually pan out!

1. I hate — hate — how Musk has tainted a perfectly good Spaceballs reference.
2. I had a fudge factor in there to account for paying taxes on the gains along the way, as some would be in a TFSA/RRSP/FHSA, some in a non-registered, but the gains are getting large enough that rough estimate likely isn’t enough, so maybe just $500k after tax.
3. Ok, a good chunk is tied up in RRSPs, so they’d need a larger mortgage in reality to actually buy a place.
4. And a rough penalty of a bit over ten thousand dollars in moving costs, but not accounting for the psychological toll of moving — that’s what we’re comparing the moving costs to.
5. I mean, whether the decision was good or not should be judged based on what was known when it was made, not the post hoc result, but after being a very, very tiny minority voice in an onslaught of housing bullishness, it’s very nice to have the post hoc result.