6. Optimize RRSP and RRIF Withdrawals

An RRSP must become a RRIF by December 31 of the year you turn 71. There’s no way around that deadline but nothing stops you from converting earlier, and for a meaningful number of retirees, earlier is exactly right.

Converting some or all of an RRSP to a RRIF before 71 unlocks two things a straight RRSP can’t offer. First, once you’re 65, RRIF withdrawals qualify as eligible pension income, opening the door to the $2,000 pension income tax credit and to splitting up to 50% of that income with a spouse neither of which is available on ordinary RRSP withdrawals before a RRIF or annuity is in place. Second, an early, voluntary RRIF withdrawal is one you’re choosing to take, at a time and amount you control, rather than one dictated later by an escalating minimum-withdrawal formula.

That formula is worth understanding in detail, because it’s the mechanism that turns a comfortable RRSP into a potential tax problem if it’s left untouched for too long. Once a RRIF is established, the government requires a minimum withdrawal each year based on a percentage of the account’s value on January 1, and that percentage rises steadily with age: 4.00% at 65, 5.28% at 71 (the age most people first face the requirement), 5.82% at 75, 6.82% at 80, and all the way up to 20% at 95 and beyond. There’s no ceiling you can always withdraw more than the minimum but there’s also no way to withdraw less, which is exactly why a large, untouched RRSP that becomes an equally large RRIF can eventually force uncomfortably big taxable withdrawals in years when you may not need the cash at all.

Consider a retiree who converts nothing until 71 and arrives at that deadline with $900,000 in an RRSP. At 71, the mandatory 5.28% minimum forces a withdrawal of roughly $47,500 in the very first year fully taxable, on top of CPP, OAS, and any pension income already being received. By 80, with a balance that has likely grown despite withdrawals, the 6.82% minimum could force a withdrawal well past $70,000, a jump that has more to do with an arbitrary birthday than with what that retiree actually needs to spend that year. Compare that to a retiree who begins modest, voluntary withdrawals at 65, drawing the account down gradually while filling up a lower tax bracket each year, and arrives at 71 with a meaningfully smaller balance and therefore a meaningfully smaller mandatory minimum for the rest of their life.

One practical note on cash flow: while the mandatory minimum itself has no tax withheld at source (though it’s still fully taxable when you file), any amount withdrawn above that minimum is subject to withholding tax 10% on the portion up to $5,000, 20% between $5,001 and $15,000, and 30% above that, outside Quebec. That withholding isn’t an extra tax; it’s a prepayment against what you’ll ultimately owe, reconciled when you file. But it does mean a large discretionary withdrawal arrives with less cash in hand than the sticker amount suggests, worth factoring into any decision to draw extra from a RRIF in a given year.

There’s a further detail worth knowing: if your spouse is younger than you, you can elect at the time the RRIF is set up, and not afterward to calculate the minimum withdrawal using their age instead of yours. Because a younger age carries a lower prescribed percentage, this single election can meaningfully reduce mandatory withdrawals for as long as the RRIF exists, particularly useful when the RRIF holder is significantly older than their spouse.

The broader lesson echoes Section 4: an RRSP or RRIF isn’t a single decision made at retirement, but an asset that benefits from active management across two or three decades. Converting early, withdrawing more than the minimum in low-income years, and using the younger-spouse election where it applies are all ways of keeping control of the account, rather than letting an aging formula make the decision for you.

7. Delay CPP and OAS: When It Makes Sense

Few retirement decisions get debated as often, or as inconclusively, as when to start CPP and OAS. Both programs offer the same basic trade: start earlier and accept a permanently smaller monthly payment, or wait and receive permanently more. Neither choice is objectively correct, it depends on health, other income, and how you personally weigh certainty against flexibility.

CPP can begin any time between 60 and 70. Starting before 65 reduces the payment by 0.6% for every month early, up to a maximum reduction of 36% at 60. Delaying past 65 increases it by 0.7% for every month, up to a maximum increase of 42% at 70. OAS works similarly but only in one direction: there’s no option to start before 65, only the choice to defer up to age 70 for a 0.6% increase per month of delay, up to 36% more.

The Cost of Starting Early, the Reward for Waiting (2026 CPP Maximums)

Age Started	Monthly CPP (2026 maximum)	Adjustment vs. Age 65
60	≈ $965	−36%
65	$1,507.65	Baseline
70	≈ $2,141	+42%

The case for delaying is really a case about longevity insurance. The “break-even age” the point where a later start catches up to an earlier one in total dollars received typically falls somewhere in the late seventies to early eighties for both CPP and OAS. Live past that age, which a great many retirees do, and delaying comes out ahead, sometimes substantially so over a retirement that stretches into the nineties. For a healthy retiree with a family history of longevity, or simply someone uncomfortable with the risk of running short of guaranteed income late in life, deferral functions as a form of insurance the government happens to sell at a fixed, guaranteed price.

The case for starting early is just as real. A retiree in poorer health, or with a family history suggesting a shorter-than-average life expectancy, may reasonably prefer the certainty of income now over a theoretical larger payment they may never live to fully benefit from. Someone who has stopped working and has no other income to bridge the gap may simply need the cash flow at 60 or 65, regardless of the long-run math. And a retiree with a large non-registered or RRSP portfolio might take CPP early specifically to reduce how much they need to withdraw from those accounts in the early years, preserving them for later.

There’s also a quieter, tax-driven argument for delaying that often gets missed in the break-even framing: delaying CPP and OAS creates room in your sixties to draw down RRSP savings at your own pace, in years that may otherwise have relatively low income, rather than adding CPP and OAS on top of RRIF withdrawals from day one. For a retiree following the withdrawal-order logic from Section 4, deferring government benefits by even a few years can be less about the benefits themselves and more about buying time to bring a large RRSP down before mandatory RRIF withdrawals and CPP/OAS all arrive at once in your seventies.

There’s no single right answer here, which is precisely why this decision deserves real thought rather than a default. Health, other income, marital status, and how you feel about risk all point toward genuinely different answers for different people and unlike most retirement decisions, this one is largely irreversible once made.

8. Prevent Old Age Security (OAS) Clawback

The OAS clawback officially the OAS recovery tax is one of the more misunderstood mechanics in Canadian retirement planning, partly because the word “clawback” makes it sound more punitive than it actually is, and partly because so few retirees realize how many ordinary financial decisions can trigger it.

Here’s the mechanism: once your net income for the year exceeds a threshold set annually at $95,323 for the 2026 tax year the government reduces your OAS payment by 15 cents for every dollar above that line. It isn’t an all-or-nothing cliff; it’s a gradual reduction that continues until your entire OAS pension has been clawed back, which for someone aged 65 to 74 in 2026 happens at roughly $154,700 of net income. Because Service Canada calculates the recovery tax using your previous year’s tax return, a high-income year doesn’t reduce your OAS immediately, it reduces the payments you receive from the following July through the June after that, exactly the kind of one-year lag that catches people off guard when income shifts unexpectedly.

What makes the clawback especially tricky to plan around is how many income sources feed into that net income figure, several in ways that aren’t obvious. A large RRIF withdrawal counts in full, dollar for dollar the single most common reason retirees find themselves in clawback territory, often because a mandatory minimum withdrawal has grown alongside their RRIF balance over the years. Capital gains count too, though only the taxable half of the gain sell a large non-registered holding in one year and you may push your net income past the threshold even though only half the gain was technically taxable. And Canadian dividends carry a quirk that regularly surprises people: because of the dividend gross-up mechanism, the amount that counts toward your net income for clawback purposes is actually higher than the cash dividend you received. An eligible dividend is grossed up by 38%, meaning $10,000 of actual dividends can add roughly $13,800 to the income figure used to test for clawback, even though the dividend tax credit later offsets much of the extra tax.

None of this needs to feel like a trap, because the tools to manage it are largely the ones covered elsewhere in this guide, simply pointed at a specific target. TFSA withdrawals do the most direct work here, since as covered in Section 5 they add no income at all, meaning a retiree who needs extra cash in a year when they’re already close to the threshold should generally reach for the TFSA before the RRIF. Pension income splitting, covered next in Section 9, can move income from the higher-earning spouse (who may be approaching the clawback zone) to the lower-earning one (who likely isn’t), since the clawback applies individually, not to household income splitting income between two people who are each below the threshold can keep OAS fully intact for both, even when the household’s combined income is substantial. Income smoothing, covered in Section 11, addresses the underlying cause directly: an evenly spread series of withdrawals is far less likely to spike past the threshold in any single year than one large withdrawal taken all at once. And simple withdrawal timing involves taking a large RRSP withdrawal or realizing a large capital gain in a year before OAS begins, rather than after sidesteps the clawback for that specific piece of income entirely.

The retirees who navigate this best tend to share one habit: they estimate their net income for the year before December 31, not after, leaving enough time to adjust a withdrawal, harvest a loss, or shift income to a spouse while it still matters.

9. Use Pension Income Splitting

Pension income splitting is one of the few tax strategies available to Canadian retirees that costs nothing to set up, requires no restructuring of accounts, and can be adjusted every year based on how income actually unfolds. And yet a surprising number of eligible couples never use it, often simply because no one told them it existed.

The mechanic is straightforward: each year, a couple can jointly elect using CRA Form T1032, filed with both tax returns to allocate up to 50% of one spouse’s eligible pension income to the other, entirely on paper. No money changes hands, no accounts are retitled, and the election can be revisited, and changed, every tax season based on that year’s numbers. Eligibility depends partly on age: before 65, only payments from a registered employer pension plan qualify; from 65 onward, the definition widens to include RRIF and RRSP annuity income as well, one more reason converting some RRSP savings to a RRIF before 71 can be worthwhile. CPP and OAS are notably excluded from this particular mechanism CPP has its own, separate sharing arrangement through Service Canada, discussed in Section 12, and OAS can’t be split at all.

The benefits stack up in ways that are easy to underestimate. The most obvious is a lower household tax bill: because Canada’s tax brackets are progressive, a couple where one spouse has $70,000 of pension income and the other has none will generally pay more combined tax than a couple splitting that same $70,000 evenly, simply because more of it stays in lower brackets. Less obvious, but often more valuable, is the effect on the OAS clawback since the recovery tax applies to each spouse’s individual net income, shifting income away from a higher-earning spouse approaching the $95,323 threshold can protect their OAS entirely, even without changing the couple’s total income by a single dollar. Splitting can also help both spouses claim the $2,000 pension income tax credit rather than just one, and it can preserve the age amount credit for a spouse whose income would otherwise be too high to claim it in full.

Consider a couple where one spouse, Margaret, receives $65,000 a year in RRIF and pension income, while her spouse, Alan, receives only CPP and OAS. Without splitting, Margaret’s income sits well into a higher bracket than Alan’s, and the couple pays combined tax as though one person earned everything. By electing to allocate roughly $30,000 of Margaret’s eligible pension income to Alan, the couple shifts a meaningful slice of that income into Alan’s lower bracket, reducing their combined tax bill by a real, calculable amount each year and if Margaret’s income had been edging toward the OAS threshold, the split could preserve OAS that would otherwise have been put at risk.

The only real cost of pension income splitting is the annual paperwork, and most tax software calculates the optimal split automatically once both spouses’ information is entered. For couples with any meaningful gap in pension income, it’s rarely a question of whether to use it only to confirm, every year, that the split is actually being filed.

10. Manage Capital Gains Efficiently

Capital gains occupy a unique place in the retirement tax picture: they’re one of the few major income sources where you largely choose when the tax bill arrives, simply by choosing when to sell.

In Canada, only half of a capital gain is included in taxable income, a rate that was the subject of considerable back-and-forth in recent years. The 2024 federal budget proposed raising that inclusion rate from one-half to two-thirds on gains above $250,000 annually, a change that was first deferred and then, in March 2025, cancelled outright by the government. As of the 2026 tax year, the inclusion rate remains at one-half, exactly where it has sat for years, which means a $100,000 capital gain still adds only $50,000 to taxable income often the lightest-taxed way to realize a large sum from a non-registered account.

That said, lightest-taxed doesn’t mean tax-free, and a large gain realized in a single year can still push a retiree into a higher bracket or toward the OAS clawback threshold just as easily as an equivalent RRIF withdrawal would. This is where deliberate timing matters. Rather than selling an entire appreciated position in one year, a retiree with flexibility can spread the sale across two or three tax years, realizing only enough gain each year to stay within a target bracket the same income-smoothing logic that applies to RRIF withdrawals, applied instead to a non-registered portfolio.

Tax-loss harvesting is the natural complement to this strategy: selling an investment currently worth less than you paid for it crystallizes a capital loss, which can offset capital gains realized in the same year, be carried back up to three years against gains already taxed, or carried forward indefinitely against future gains. A retiree planning a large, necessary sale in a given year can often reduce its tax cost meaningfully by first reviewing the rest of the portfolio for any losses worth realizing alongside it. The one rule worth knowing well is the superficial loss rule: buying back the same or an identical investment within 30 days before or after the sale disqualifies the loss for tax purposes, so any repurchase needs to wait out that window, or use a genuinely different, if similar, holding.

All of this applies specifically to non-registered investments inside an RRSP, RRIF, or TFSA, capital gains and losses simply don’t exist for tax purposes, since withdrawals from registered accounts are taxed as ordinary income, or not taxed at all in the TFSA’s case, regardless of what generated the growth inside. That’s precisely why asset location matters as much as asset selection: investments most likely to generate capital gains are often better held in a non-registered account, where that favourable tax treatment can actually be used, rather than inside an RRSP, where a capital gain and a bond’s interest income end up taxed identically once withdrawn.

For retirees with a meaningful non-registered portfolio, capital gains management deserves the same annual attention as RRIF withdrawals: a year-end review of what’s appreciated, what’s declined, and how much room remains before the next bracket or threshold, rather than a decision made only when a sale becomes unavoidable.

11. Minimize Taxes Through Income Smoothing

Income smoothing is less a single technique than a mindset that ties together nearly everything covered so far: rather than reacting to withdrawals year by year, a retiree who smooths income treats their remaining accounts as one pool to be drawn down evenly over their expected retirement, deliberately avoiding the large, lumpy withdrawals that push income into brackets it didn’t need to reach.

The logic is easiest to see with a number. Suppose a retiree needs $400,000 from an RRSP over the next ten years and has no other pressing need for the money sooner. Withdrawing it as two $200,000 lump sums perhaps once at 65 and once at 71 would push a significant portion of each withdrawal into the higher federal brackets, since $200,000 in a single year lands well past the $181,440 and even the $258,482 thresholds where the top rates apply. Spread across ten years instead, at $40,000 annually, that same $400,000 might stay comfortably within the second or third federal bracket the entire time, depending on the retiree’s other income often saving several thousand dollars in combined federal and provincial tax over the decade, without changing a single investment decision, only the timing of withdrawals.

The same logic works in reverse for RRIF minimums. A retiree who lets an RRSP grow untouched until 71 is choosing, whether they realize it or not, to defer all of that income into years when the mandatory minimum withdrawal percentage is higher and climbing: 5.28% at 71, versus 4.00% at 65. Beginning modest, voluntary withdrawals years before the RRIF conversion deadline is itself a form of income smoothing it trades a smaller, avoidable tax bill today for a meaningfully smaller mandatory bill in your eighties and nineties, when the percentage, and often the account balance itself, would otherwise both be larger.

Smoothing also pays off around one-time events that are easy to see coming: the year you sell a rental property, the year a large non-registered gain needs to be realized, or the year before CPP and OAS begin, when income is often at its lowest point in retirement. A retiree who recognizes a coming low-income year the gap between stopping work and starting government benefits, for instance has a genuine opportunity to accelerate RRSP withdrawals or realize gains specifically because that year’s bracket has more room than most years will.

None of this requires precision forecasting. It requires a rough annual estimate of what my income will look like this year, and is there room below the next bracket or the OAS threshold worth using deliberately repeated every December for as long as the accounts last. Retirees who make this a habit rarely save a fortune in any single year, but a few thousand dollars saved consistently across two or three decades adds up to a genuinely different retirement than one drained by a handful of large, badly timed withdrawals.

12. Tax Planning for Couples

Every strategy in this guide becomes more powerful, and often more complicated, once a spouse enters the picture. Couples have planning tools unavailable to single retirees, and the best-run retirement plans tend to treat both spouses’ finances as a single, coordinated system rather than two separate accounts that happen to share a household.

Pension income splitting, covered in Section 9, is the most direct of these tools, and it’s worth distinguishing clearly from CPP pension sharing, a separate program entirely. Where pension splitting is an annual tax-return election covering RRIF, RRSP annuity, and DB pension income, CPP sharing is an application filed directly with Service Canada that physically redirects a portion of each spouse’s actual CPP payment to the other, based on the years they lived together during their contributory periods. Both spouses generally need to be 60 or older, living together, and either receiving or eligible for CPP. Because the two programs cover different income pension income versus CPP specifically and work through different mechanisms, a tax election versus an ongoing payment reassignment many couples benefit from using both at once rather than choosing between them.

TFSA optimization for couples adds another layer: since contribution room is individual, not shared, a couple who have both been eligible and contributing room since the TFSA’s 2009 launch can treat their two accounts as a combined $218,000 reservoir of tax-free, benefit-neutral withdrawal room between them often directing withdrawals through whichever spouse’s TFSA leaves the healthier RRSP or RRIF balance intact for the other, or whichever spouse is closer to an OAS clawback or GIS threshold that a taxable withdrawal would jeopardize.

Staggering RRIF withdrawals between spouses is a natural extension of income smoothing applied at the household level: rather than both spouses drawing their RRIFs down proportionally every year, a couple can direct larger withdrawals through whichever spouse has more bracket room available in a given year, effectively using two tax returns’ worth of lower brackets instead of one. The same logic extends to coordinating CPP start dates a couple doesn’t need to start CPP at the same age, and a couple where one spouse defers to 70 while the other starts earlier can balance guaranteed lifetime income against near-term cash flow in a way that suits their specific ages, health, and spending needs, rather than defaulting to a single “right” answer for both.

Estate considerations round out the picture. A surviving spouse can generally roll over a deceased spouse’s RRSP or RRIF without any immediate tax, deferring the eventual tax bill rather than triggering it at death but that rollover also means a surviving spouse often ends up with a single, larger RRIF than either spouse held individually, facing correspondingly larger mandatory minimum withdrawals alone. Couples who plan ahead sometimes deliberately draw down the RRSP or RRIF of whichever spouse has the shorter life expectancy a sensitive calculation, but a real one somewhat faster during their lifetimes, specifically to reduce the size of that eventual combined RRIF and the tax burden it will eventually place on a surviving spouse filing alone.

13. Tax Planning for Self-Employed Canadians

Self-employed and incorporated Canadians face a version of retirement tax planning that looks quite different from the CPP-and-RRSP framework most of this guide has covered, largely because a corporation opens up planning tools that simply don’t exist for an employee drawing a T4 salary.

Individual Pension Plans (IPPs)

An Individual Pension Plan is the clearest example. An IPP is a defined benefit pension plan sponsored by a corporation for a single high-earning employee, typically the owner, and unlike an RRSP’s fixed 18%-of-income formula, IPP contribution room is calculated actuarially, based on age, salary, and years of service which means it grows meaningfully larger than an RRSP’s as the owner ages. Around 40, the two are roughly comparable; by 50, an IPP can typically accommodate something in the range of $10,000 to $11,000 more per year than the RRSP maximum; by 60, that gap often exceeds $19,000 to $20,000 annually, and past service can sometimes be purchased retroactively to 1991 for a substantial one-time corporate deduction. The trade-off is real, though: IPP assets are generally locked in, without the RRSP-to-RRIF flexibility to withdraw a lump sum on your own schedule, and the plan carries higher actuarial and administrative costs. For an incorporated professional over 45 or 50 with a T4 salary well above the RRSP maximum and a genuine interest in maximizing tax-deferred savings rather than preserving flexibility, an IPP is worth a serious look; for almost anyone younger or with more modest income, the RRSP typically remains the simpler, cheaper choice.

Corporate Investment Accounts

Corporate investment accounts bring their own, sharper-edged trade-off. The small business deduction, which lets a Canadian-controlled private corporation pay a reduced tax rate on up to $500,000 of active business income, begins to shrink once the corporation’s passive investment income interest, portfolio dividends, and taxable capital gains held inside the company exceeds $50,000 in the prior year. The business limit drops by $5 for every $1 of passive income above that threshold, disappearing entirely at $150,000. To put a number on it: a corporation earning $70,000 of passive investment income in a year not unreasonable for a well-established professional practice with a seven-figure portfolio sits $20,000 above the threshold, which reduces the following year’s business limit by $100,000, from $500,000 down to $400,000. Losing access to the small business rate on that $100,000 of active income typically costs tens of thousands of dollars a year in additional corporate tax, which is why many incorporated professionals treat $50,000 of passive income as a real ceiling to plan around, not just a technical threshold.

Dividends vs. Salary in Retirement

The choice between salary and dividends threads through nearly all of this. Salary is deductible to the corporation and creates both CPP contribution history and RRSP room, but it requires paying both the employer and employee halves of CPP. Dividends avoid CPP contributions entirely and are taxed personally at a lower effective rate thanks to the dividend tax credit, but they build no CPP entitlement and no RRSP room whatsoever. Many incorporated owners lean toward salary earlier in their career, while RRSP room and CPP contribution history still matter, and shift the balance toward dividends later, once both are largely built and the priority turns to extracting income as efficiently as possible.

Holding Companies

The passive-income ceiling described above is exactly why many incorporated professionals eventually separate their operating business from their investment portfolio, and that separation is usually achieved through a holding company. Rather than investing surplus cash inside the operating company (Opco), profits are paid up to a separate holding company (Holdco) as an inter-corporate dividend, which is generally tax-free between connected corporations. The Holdco can then hold investments, real estate, or life insurance without contaminating Opco’s passive-income test, while also offering a genuine creditor-protection benefit: assets moved to the Holdco are generally shielded from claims against the operating business, a real consideration for anyone in a higher-liability profession or industry.

Business Succession Planning

Business succession and the eventual sale of a qualifying business bring one more meaningful planning tool: the lifetime capital gains exemption, which in 2026 shelters up to roughly $1.275 million of gain on the sale of qualified small business corporation shares, or qualifying farm and fishing property, from tax entirely. Structures like an estate freeze exchanging growth shares for fixed-value preferred shares and allowing future growth to accrue to children or a family trust and “purifying” a corporation of excess passive assets to preserve its qualification for that exemption are specialized enough to warrant a dedicated conversation with an accountant, but retiring business owners who leave this planning until the year they sell routinely leave a meaningful, avoidable tax bill on the table simply for lack of lead time.

Tax-efficient retirement planning was never really about outsmarting the system. It’s about understanding it clearly enough to make deliberate choices which account to draw from, when to start a benefit, how to split income with a spouse rather than defaulting to whatever feels simplest in the moment. None of the strategies in this guide work in isolation forever; tax rules shift, thresholds get indexed, and personal circumstances change every year of retirement. The habit worth building isn’t memorizing this year’s numbers, but returning to this kind of thinking annually, for as long as retirement lasts.