Blessed by the Potato

A common myth is that you can’t withdraw from your RRSP until retirement (except for programs like the HBP or LLP). There’s actually no such restriction: you can withdraw at any time, but the withdrawal will get added to your income for the year, and you will lose that room. Now, most of the time that means you’re going to want to wait until retirement before withdrawing anyway.

But what if you’re in a period of your life where your income is temporarily low? Mat/pat leave, or a job loss, for example? It may make sense then to pull some of your money or investments out of the tax shelter when you can do so at a lower tax rate.

Reminder: The RRSP is beneficial in two basic ways: it provides tax-free compounding of your investments, and lets you contribute with pre-tax money, so you can engage in tax arbitrage by deferring the tax until later, when you might be in a lower tax bracket. Much of the time the tax arbitrage is not much of a benefit, as clawbacks of various programs can make your effective tax rate in retirement higher than raw income might suggest. But, if you have a year with minimal income (because of job loss, mat leave, etc.) then you may find yourself in the lowest possible tax bracket (lower than when you contributed the money) so there’s an advantage to withdrawing then. But you’ll be trading away the one benefit for the other. Is it worth it?

There are a few basic scenarios to consider, and your own different tax rates and rates of return will play a role, too.

The biggest question to start with is whether or not you have excess RRSP room. For many younger people who start saving early, getting 18% of your pay in RRSP room and a chunk of TFSA room (~10% of the YMPE) means that you’ll have more tax shelter space than you will fill. In that case, burning some of it up to withdraw in a low-income year is an easier decision: you pay tax at a lower rate, then put it back in when you’re in a higher bracket. The benefit from tax-arbitrage just between the bottom rate of 20% and a middling ~30% is a one-time gain of ~10%, which is going to far exceed one or two years of tax on investment growth (assuming you don’t actually need the money to pay for your expenses while out of the workforce).

If you have excess RRSP room, it’s a pretty easy choice to withdraw during a lower-earning year. You can even plan in advance for this where possible. For instance, you may ordinarily use your TFSA for long-term savings, but if you’re pregnant and expecting to be off work the following year, you might choose instead to use your RRSP (or even withdraw some funds from your TFSA to contribute to your RRSP), then withdraw those funds when your earnings are lower the next year.

We’re not all in that boat though — if you started building up a non-registered account before the TFSA was introduced, or if your RRSP room is used up by a pension, or if you have earnings that don’t generate RRSP space (like grad school stipends or dividends from your small business), or if you just simply have a high savings rate (which may be needed for early retirement plans or because you wasted your 20s in grad school and have to play catch-up or because you’re renting-and-investing-the-difference), then you might not have any available RRSP contribution room to put back what you take out in a low-income year. In that case, you’ll want to be more careful and certain that a withdrawal in a low-income year will really help you before you go through with it and burn the space.

Now it’s harder. You could get the one-time benefit of pulling money out at a low rate, but then you’re going to have non-registered investments that grow more slowly due to the tax drag than registered ones — and if you expect to be in a low bracket at retirement anyway (or for several more years as your disability takes time to resolve), then taking the money out early is of no real benefit to you.

There are going to be a number of factors at play: you’ll have a higher rate of compounding on your RRSP investments than non-registered ones (made even more complicated by the fact that some capital gains can be deferred a long time even in a non-registered account). The length of time to consider will matter, too: if you’ll have to pull the money out in a few years anyway, then it may make sense to withdraw early, as the difference in growth rates may not add up to much compared to the difference in effective tax rates of the withdrawal. If it would be a few decades before you would otherwise take money out of your maxed-out RRSP, then even a modest tax drag can add up to a large effect and you’d be better off just leaving things alone in your off year.

To help look at a few scenarios, I’ve put together a spreadsheet available here. You can play around with it to see which is better, and also how much you’d be off by in different scenarios if your assumptions are wrong.

For example, we can run a scenario where Wayfare is 37 and would be in the 20% tax bracket this year, but in a few years is able to return to work and get back to the 30% tax bracket, and then will stay in that bracket until the money is needed in 18 years (retirement at 55). Let’s assume her investments grow at 7% inside the RRSP, and at 6.5% outside. In that case, she is a bit better off to withdraw early (~$122 extra in the future for $1000 withdrawn today). If instead the disability is permanent, and the later withdrawal would also be at 20%, then she would have been better off leaving the money in her RRSP. The optimism costs ~$193 (in lost growth in future dollars) for $1000 withdrawn today (about 8% of her future investment value). If her non-registered investments are less tax-efficient and only grow at 5.9% (because her earnings power returns, for example), it’s an even worse deal to withdraw early.

It’s a tough call — you have to know the future to say for sure how much (if any) you’ll save.

And for the case of someone with no spare RRSP room and non-registered investments, there’s a similar dilemma of whether to realize the gains now in a low bracket, paying tax now so you have less to continue investing, but resetting your cost basis higher for the future. The second tab of that spreadsheet looks at some cases for that decision.

Of course, we’re actually facing these decisions. While going in I thought it would make sense to take advantage of a low-income year, after looking at a few scenarios we’re not going to burn any RRSP room as long as there’s still non-registered investments standing behind the emergency fund.