A CAD 400,000 gap in how Canada taxes a RRIF at death looks like the whole answer, until you ask how much of that RRIF you will actually have left, and where you will genuinely be living when it matters.

Picture two Ontario retirees who started in nearly identical positions. Both retired at 55 with a RRSP worth about CAD 1.5 million. Both are now in their late eighties, widowed, with no spouse left to roll a RRIF over to. One still lives in Ontario. The other genuinely moved away years ago and has been a non-resident of Canada ever since.

If each of them dies this year with a CAD 1.5 million RRIF still sitting there, the Canadian tax difference between them can be roughly CAD 400,000. The resident’s RRIF is generally included in income on the final return, pushing much of a large balance into Canada’s highest marginal tax brackets. For a genuinely non-resident annuitant, the best-supported Canadian treatment is instead a deemed RRIF payment subject to Part XIII withholding, normally at 25 percent.

That number is genuinely startling, and it is tempting to stop there and call it the answer. We are not going to stop there, because the CAD 1.5 million RRIF at death is not a starting assumption. It is an outcome. To get there, each retiree had to make three decades of decisions: how much to withdraw every year, whether to melt down the RRSP deliberately or leave it mostly alone, whether the money they withdrew was spent or quietly reinvested somewhere else, when to actually leave Canada if they left at all, what happened to their CPP and Old Age Security along the way, whether they gave up pension income splitting with a spouse, what the destination country did with its own tax system, and, somewhat inconveniently, whether they lived long enough to carry out whatever plan they had made.

This piece started as a fairly ordinary snowbird-versus-expat tax comparison for Sovereign Canadian readers. It turned into something closer to a lifecycle RRSP and RRIF problem, and the answer that came out the other end is less tidy than “leave” or “stay.” The answer turns on how much registered wealth you are likely to have left when you die, and where you will genuinely be a resident when that happens.

Snowbirding and Emigrating Are Not the Same Thing

Before any of the tax mechanics make sense, it helps to pull apart four things that get casually lumped together: Canadian citizenship, immigration status in another country, Canadian tax residence, and provincial health insurance residence.

Citizenship does not change when you spend your winters elsewhere. A Canadian citizen who spends five months a year in Arizona is still a Canadian citizen, full stop, and nothing about tax residence touches that.

Immigration status abroad is a separate question again. Getting a temporary resident visa, a retirement permit, or even a long-stay residence card in another country does not automatically make you a non-resident of Canada for tax purposes. Plenty of Canadians hold foreign residence permits while remaining squarely Canadian tax residents, because they kept a home, a spouse, and an ordinary life back in Canada.

Canadian tax residence is a question of fact, not a simple day count. The Canada Revenue Agency’s guidance on this, set out in Income Tax Folio S5-F1-C1, looks at residential ties: a dwelling available to you in Canada, a spouse or common-law partner who stays behind, dependants, and the general pattern of where you actually live your life. The CRA’s separate guidance for Canadians leaving the country makes the same basic point: keeping significant Canadian residential ties will normally keep you within the Canadian tax-residence system.

A Canadian who spends six months a year in Florida but keeps an available family home and a spouse in Ontario may therefore still be a factual resident of Canada, taxed here on worldwide income, regardless of how long the Florida stay was. Genuinely ceasing to be a Canadian resident generally means substantially severing those ties and establishing an ordinary life somewhere else rather than treating it as an extended vacation. Where someone has residential ties in two countries at once, a tax treaty’s tie-breaker rules can decide the question, but that is a topic for a cross-border tax advisor, not something to sort out casually.

Provincial health insurance residence is its own fourth category, governed by its own separate rules.

The practical point is that someone can spend months abroad every year and remain fully Canadian resident for tax purposes, which is what most snowbirds are doing. And someone can qualify for a foreign residence visa without that visa, on its own, making them a Canadian non-resident. The two questions move independently of each other, and conflating them is where a lot of casual retirement advice goes wrong.

For readers considering the second path rather than simply a winter escape, the broader Sovereign Canadian Expat Living library looks at what actually establishing a life in another country entails.

What a Canadian Snowbird Keeps

A snowbird, in the sense used here, is a Canadian who spends a meaningful chunk of the year outside the country but remains a Canadian tax resident. That status comes with a specific bundle of things, and it is worth being explicit about what is actually in the bundle before weighing it against the alternative.

A resident pays ordinary Canadian progressive tax on worldwide income, with the full set of federal and Ontario, or other provincial, brackets and the usual credits where eligible. RRSP and RRIF withdrawals are taxed as ordinary income, the way they always have been, with no special non-resident withholding involved.

Where a spouse or common-law partner is also Canadian resident, eligible pension income can be split between them under section 60.03 of the Income Tax Act, which can meaningfully lower a couple’s combined tax bill when one spouse’s retirement income is much larger than the other’s. CPP and OAS continue without the additional non-resident withholding and treaty questions that arise once someone moves abroad. TFSA growth and withdrawals stay fully tax-free in Canada, which is not something every destination country respects once you actually move there.

And, for an Ontario resident specifically, provincial healthcare under the Ontario Health Insurance Plan continues, provided the residency rules are satisfied.

Those rules are worth getting precisely right, because the article you have probably read somewhere else is wrong on the margins. Ontario Regulation 552 requires an ordinary resident to remain physically present in Ontario for at least 153 days in any given 12-month period, while continuing to maintain Ontario as their primary place of residence. Ontario’s own OHIP guidance for time spent outside Canada confirms the practical rules.

That works out to a little over five months. Because the statutory test is framed as any given 12-month period rather than simply a calendar year, building a recurring snowbird schedule around exactly 153 days leaves very little room for error. Aiming for something closer to 160 or 170 days gives substantially more breathing room.

There is a separate, more limited provision for longer absences. Regulation 552 allows a qualifying resident who met the 153-day requirement in each of the two consecutive 12-month periods immediately before leaving to be treated as satisfying the physical-presence requirement for a maximum of two 12-month periods while travelling outside Ontario. The regulation also says that this exemption cannot be used again until the person has accumulated at least 153 qualifying days in each of at least five consecutive 12-month periods.

In plain terms, this is an occasional allowance for a genuinely longer absence, not a mechanism for spending seven or eight months abroad every single winter indefinitely. Anyone building a retirement plan around routinely being away more than about seven months a year should not assume OHIP survives that pattern.

Provincial coverage also does not travel well. Ontario currently pays no more than CAD 50 per day for qualifying emergency outpatient hospital services outside Canada and generally CAD 200 per day for inpatient services, rising to CAD 400 for certain higher-acuity hospital settings. Ontario itself recommends private health insurance before leaving.

For a broader look at what Canadians are actually preserving when they maintain provincial healthcare coverage, see Is Canadian Healthcare Actually World Class?.

The Underappreciated Retirement Window: About 55 to 70

The years between roughly 55 and 70 do something unusual to a retiree’s tax situation, and it is worth spending a section on why, because it sets up everything that follows.

Someone who retires at 55 has typically stopped earning employment income, has not yet started CPP, available as early as 60 but often delayed, has not yet started OAS, available at 65 at the earliest, and has no RRIF minimum withdrawal obligation unless they have already converted part of the RRSP to a RRIF. An RRSP itself must be dealt with by the end of the year in which the owner turns 71.

For a decade and a half, in other words, a retiree can have remarkably little taxable income unless they manufacture some.

This is the foundation of the conventional RRSP meltdown idea: deliberately withdraw RRSP money during these low-income years, pay tax at relatively gentle Canadian marginal rates, and reduce the RRIF balance that will eventually be subject to mandatory withdrawals, CPP, OAS, and potentially the OAS recovery tax all stacking on top of each other later.

It is a reasonable idea, and the arithmetic behind it is real.

It is worth being clear, though, that withdrawing the maximum possible amount is not automatically the right amount, and the reason has to do with what happens to the money once it leaves the RRSP.

Every dollar pulled out loses the tax-deferred compounding it would otherwise have enjoyed inside the registered plan. If that dollar is spent, the comparison is straightforward: you paid tax on it at a known rate and used it for something. If it is not spent, it has to live somewhere else, typically a TFSA if room remains or a non-registered account where future income and gains face a different tax regime.

That drag matters, and it is one of the places where simplistic meltdown advice tends to go wrong by treating every early withdrawal as costless reinvestment. It turns out to be one of the most important variables in the entire question, as the modelling further down makes concrete.

What Actually Changes When You Become a Non-Resident

Genuinely ceasing Canadian tax residence triggers a specific set of consequences, and it is worth walking through them cleanly before layering treaty complexity on top.

The departure date itself matters because Canada applies what is generally called departure tax: a deemed disposition of many assets at fair market value immediately before you cease to be resident, as though you had sold them and immediately bought them back.

The CRA’s departure-tax guidance sets out the mechanics in detail, but the point that matters most for this discussion is a specific exclusion: RRSPs and RRIFs are excluded from the deemed-disposition regime. So are TFSAs, although a TFSA’s exclusion from Canadian departure tax does not mean the destination country will treat the account as tax-free after you move there.

Canadian real estate is also excluded from the ordinary departure-tax deemed disposition, but remains inside the Canadian tax system and raises its own separate questions around rental withholding, reporting and eventual sale.

Once you are a non-resident, Canadian-source RRSP and RRIF withdrawals are normally subject to Canadian non-resident withholding tax under Part XIII, generally at 25 percent unless a treaty reduces the rate.

A non-resident can also elect under section 217 of the Income Tax Act to have certain Canadian-source retirement income taxed through a Canadian return. The CRA specifically lists most RRSP and RRIF income, CPP, OAS and many pension benefits as income potentially eligible for the election.

This is genuinely useful in some cases and genuinely misunderstood in many descriptions of it.

A section 217 election is not simply resident taxation with a different label. The calculation brings worldwide income into the picture, and worldwide income can also affect the non-refundable credits available. CRA explains those mechanics in its section 217 guide and its Statement of World Income guidance.

The election also does nothing to restore ordinary pension income splitting, which generally remains unavailable to factual non-residents.

Where a tax treaty already caps Canadian withholding below 25 percent for the income in question, that treaty treatment can be more important than section 217. Which brings us to the next complication.

Treaty Countries Can Change the RRIF Arithmetic, Within Limits

The headline fact that gets repeated constantly in expat retirement content is that Canada’s tax treaty with the United States caps Canadian tax on qualifying periodic pension payments to a US resident at 15 percent instead of the default 25 percent.

That fact is true. It is also routinely overstated into something it is not: an invitation to simply empty a RRIF at 15 percent whenever convenient.

Article XVIII of the Canada-US tax treaty limits Canadian tax to 15 percent where the US resident is the beneficial owner of a periodic pension payment.

That word periodic matters.

Canada’s Income Tax Conventions Interpretation Act specifically defines what counts as a periodic pension payment for treaty purposes. For a RRIF, the test generally allows payments up to a ceiling based on the greater of twice the applicable RRIF minimum or 10 percent of the fund’s beginning-of-year value, subject to the detailed statutory calculation. Once payments exceed the statutory periodic-payment threshold, the excess does not simply inherit the treaty’s periodic rate.

A payment in full or partial commutation of an RRSP retirement income is also expressly excluded from the periodic-payment definition.

This is why the distinction between an RRSP and a RRIF can matter enormously for a non-resident drawing registered assets.

It also matters that treaty rates genuinely differ by country. The 15 percent figure most commonly quoted for the United States is not universal. Mexico’s treaty with Canada, for example, also limits Canadian tax on qualifying periodic pension payments to the lesser of 15 percent of the gross amount and a specified resident-rate comparison.

Spain likewise has treaty protection for periodic pension payments. France’s treaty instead gives Canada taxing rights over Canadian-source pensions covered by its pension article without imposing the same 15 percent periodic-payment ceiling. Thailand’s pension article also permits Canadian taxation without a comparable percentage cap for pensions within that article.

Japan is different again. The Canada-Japan treaty has no general private-pension article equivalent to the Canada-US provision, leaving income not otherwise dealt with to its other-income article, which permits source-country taxation.

None of this means those countries are good or bad retirement destinations. It means the common shorthand that “moving abroad gets you a 15 percent RRIF rate” is wrong.

The result depends on the country, the treaty, the type of registered plan, the character of the payment and its size.

A non-resident’s Canadian withholding, whatever the rate, is also not the full story of what you will actually pay. We come back to that when we look at destination-country taxation.

Leaving Canada at 55 Has an OAS Cost

Old Age Security is funded differently from CPP, and the mechanics catch a lot of people by surprise.

A full OAS pension generally requires 40 years of Canadian residence after age 18. Someone with fewer than 40 qualifying years can receive a partial pension calculated at one-fortieth of the full pension for each qualifying year.

Service Canada’s current OAS benefit guidance also explains the separate rule that matters after emigration: to continue receiving OAS while living outside Canada, a person generally needs at least 20 years of Canadian residence after turning 18, although a social security agreement can sometimes help establish eligibility.

That last point is worth being precise about.

A social security agreement can help someone meet the minimum residence requirement necessary to qualify. It does notconvert those foreign years into additional Canadian years for purposes of calculating the amount of the pension. Service Canada’s international-benefits guidance illustrates the distinction explicitly.

So, as a practical example, someone who lived continuously in Canada from age 18 and genuinely emigrated at 55 would have accumulated roughly 37 years of Canadian residence, giving them approximately 37/40 of a full OAS pension, assuming the other eligibility requirements were met.

That is not a universal number. Someone who immigrated to Canada later and then emigrated again at 55 could have materially fewer years and a correspondingly smaller pension.

Deferring OAS does not solve the residence-years issue. Delaying OAS beyond 65 increases the monthly payment by 0.6 percent for each month of deferral, up to 36 percent at age 70, but years spent living abroad do not become additional years of Canadian residence simply because the person waits longer to apply.

The deferral bonus and the residence-based fraction are separate calculations.

Pension Splitting Is More Valuable Than It Looks

Pension income splitting gets treated as a minor administrative footnote in most retirement planning content, and that undersells it.

For a couple with meaningfully different retirement income levels, it can be one of the more valuable things ordinary Canadian residency provides.

CRA’s current pension income splitting guidance requires both spouses or common-law partners to be residents of Canada on December 31 of the tax year, or on the date of death where applicable.

A factual non-resident cannot simply elect under section 217 and recover pension splitting. Section 217 changes how eligible Canadian-source income is taxed; it does not turn an ordinary non-resident into a Canadian resident for the pension-splitting election.

There is a narrow distinction for qualifying deemed residents. CRA’s tax guide for non-residents and deemed residents expressly contemplates pension splitting for spouses who are deemed residents of Canada at year-end. That is a specific tax-residence category, not a general exception for ordinary emigrants.

CPP and OAS themselves are not eligible pension income for this particular election. Eligible RRIF, pension and certain annuity income can be.

The value of splitting scales with how lopsided a couple’s income is, not simply with the size of their combined RRIF.

Two spouses with similar RRIF balances and similar pension entitlements gain relatively little from splitting because there is not much income imbalance to smooth out. A couple where one spouse holds most of the registered assets can gain considerably more.

In illustrative modelling for this piece, a RRIF holder with roughly CAD 150,000 of annual RRIF income splitting eligible pension income with a spouse who otherwise had little pension income saved in the neighbourhood of CAD 15,000 a year in combined household Canadian tax under the assumptions used.

That is not a universal tax saving. But it illustrates why a lopsided-income couple should put an actual value on pension splitting before assuming that a lower non-resident withholding rate automatically leaves them ahead.

Then Comes the Number That Changes the Analysis: The RRIF at Death

Everything up to this point has been about annual tax treatment while someone is alive.

What happens to a RRIF at death can be a substantially larger number than any one year’s tax comparison.

Start with the resident case.

Under subsection 146.3(6) of the Income Tax Act, when the last annuitant of a RRIF dies, the annuitant is generally deemed to have received immediately before death an amount equal to the fair market value of the RRIF. CRA describes the same general rule in its guidance on the death of a RRIF annuitant.

For a Canadian resident with no available rollover, that amount is generally included in income on the final return.

For a large RRIF, that means progressively taxing a very large amount in a single year, pushing substantial portions of the account into Ontario’s highest tax brackets.

Using illustrative 2026 federal and Ontario parameters, the modelling for this article produced the following approximate results. These are model outputs, not CRA quotations or personalized tax calculations. Other income in the year of death would change the resident figures, and the non-resident column represents Canadian tax only, before any destination-country tax:

RRIF value at death	Approximate Ontario resident tax	Non-resident Canadian tax	Approximate Canadian tax difference
CAD 500,000	CAD 245,000	CAD 125,000	CAD 120,000
CAD 1,000,000	CAD 513,000	CAD 250,000	CAD 263,000
CAD 1,500,000	CAD 780,000	CAD 375,000	CAD 405,000
CAD 2,000,000	CAD 1,048,000	CAD 500,000	CAD 548,000
CAD 3,000,000	CAD 1,583,000	CAD 750,000	CAD 833,000

That is the CAD 400,000 number from the beginning of this article.

The non-resident column reflects the best-supported Canadian treatment of the deemed RRIF amount when the last annuitant is genuinely non-resident: Part XIII applies to Canadian RRIF amounts paid or deemed paid to a non-resident, with the domestic rate normally 25 percent unless treaty relief applies.

The important treaty wrinkle is that the special lower rates discussed earlier generally apply to periodic pension payments. The Income Tax Conventions Interpretation Act expressly excludes certain non-periodic and oversized payments from that definition, while the Income Tax Act deems the RRIF’s fair market value received immediately before the last annuitant’s death.

On that basis, 25 percent is the conservative and best-supported Canadian rate to use for the terminal comparison rather than assuming the 15 percent periodic-payment rate survives at death.

But this is exactly the kind of distinction where a retiree contemplating a seven-figure estate strategy should obtain current professional cross-border tax advice before deliberately structuring their residence around it.

And remember what the table does not show: tax imposed by the country where the retiree or heirs actually live.

That can materially reduce or eliminate the apparent Canadian advantage.

The Spouse Changes the Timing

For most married or common-law couples, the RRIF-at-death comparison above is not actually a first-death issue.

Canadian rules provide several ways for qualifying RRIF amounts to pass to a surviving spouse or common-law partner on a tax-deferred basis. CRA’s RRIF-at-death guidance explains the successor-annuitant and designated-benefit mechanisms, and special transfer procedures can also apply in cross-border cases.

That means the large terminal-tax comparison usually belongs at the second death, not the first, where an eligible spousal rollover is available.

A couple’s planning horizon for this question is therefore realistically the survivor’s lifetime, not either individual spouse’s lifetime on its own.

That reframes the whole exercise.

The relevant RRIF balance is not what either spouse has today. It is what the surviving spouse is likely to still be holding at the second death, after decades of spending, investment returns, minimum withdrawals, CPP, OAS and potentially very different tax circumstances.

So Should a Wealthy Retiree Stop Melting Down the RRSP?

Here is the question that pulled this whole project away from being a straightforward tax comparison.

If a resident might eventually have a large RRIF included on a terminal return at very high marginal rates, while a genuine non-resident may face a 25 percent Canadian Part XIII rate on the terminal amount, the obvious next thought is to leave as much as possible sitting inside the RRSP or RRIF, avoid the meltdown entirely, and move abroad later.

An earlier pass of the modelling behind this piece appeared to support exactly that conclusion.

Strategies that preserved the registered account and then emigrated later frequently looked very strong.

A closer version of the modelling changed the conclusion.

The problem was spending.

The earlier model effectively allowed strategies to withdraw money and reinvest the surplus rather than forcing each strategy to fund the same actual retirement lifestyle. That can make a low-withholding withdrawal strategy look artificially powerful: the retiree pulls money from the RRIF at a favourable rate and then simply moves the money into another account.

Once the comparison was rebuilt around consistent real annual spending, something different showed up.

The stronger later-emigration strategies generally were not pure preservation strategies. They involved moderate withdrawals before emigration, with actual spending funded along the way. Simply taking the legal minimum and waiting indefinitely to leave was not generally the strongest approach.

A short note on the modelling itself: the figures below use illustrative 2026 federal and Ontario tax parameters and assume a constant 3 percent real investment return. They are designed to isolate variables, not predict anyone’s actual retirement. Future investment returns and tax law will differ.

Foreign destination-country tax is excluded from these particular results unless explicitly stated, so these are Canadian-side comparisons, not promises about total after-tax wealth.

A 55-year-old single retiree with a CAD 500,000 RRSP, spending CAD 30,000 a year in real terms and modelled to death at 90 produced the following result: an optimized resident drawdown left an estate of roughly CAD 207,000 after the Canadian taxes modelled. A strategy involving modest withdrawals followed by genuine emigration at 70 produced roughly CAD 138,000. An age-80 relocation scenario produced roughly CAD 184,000.

At this balance and under these assumptions, relocation lost to staying resident.

Now increase the starting RRSP to CAD 1.5 million and spending to CAD 60,000 a year.

The optimized resident drawdown produced roughly CAD 342,000. A modest-withdrawal strategy followed by genuine emigration at 70 produced roughly CAD 466,000, an advantage of about CAD 124,000. A separate fixed-consumption strategy involving emigration at 80 produced roughly CAD 446,000, about CAD 104,000 more than the resident benchmark in that test.

That age-80 result should not be read as evidence that taking only minimum withdrawals until 80 is optimal. The broader modelling found that moderate pre-emigration withdrawals produced better results than pure minimums at larger balances.

At this balance, later relocation could win under the Canadian-tax assumptions used.

But destination-country tax has still not entered the model.

[INTERACTIVE CALCULATOR PLACEHOLDER — Canadian Tax Comparison: Snowbird vs Non-Resident]

A future version of this calculator should let readers enter their own RRSP or RRIF balance, expected annual spending or withdrawals, CPP, OAS and other pension income, an assumed age of relocation, and an estimate of how much RRIF they expect to still be holding at death.

The important label should be Canadian Tax Comparison, not “Relocation Savings Calculator.”

Unless the tool explicitly models the foreign jurisdiction too, it cannot tell you how much moving abroad actually saves.

Spending Matters More Than the Headline RRSP Balance

The single most important variable in all of this may not be today’s RRSP balance.

It is how much of that balance will actually still be there, unspent, when the person holding it dies.

This is worth making concrete with the same CAD 1.5 million single-retiree case, varied by annual spending:

Annual spending	Resident drawdown estate	Advantage of emigrating at 70	Advantage of emigrating at 80
CAD 30,000	CAD 2,010,000	CAD 379,000	CAD 235,000
CAD 40,000	CAD 1,450,000	CAD 319,000	CAD 225,000
CAD 50,000	CAD 890,000	CAD 248,000	CAD 201,000
CAD 60,000	CAD 342,000	CAD 124,000	CAD 104,000

Again, these are outputs from the illustrative model used for this article, before destination-country tax, not general forecasts.

A retiree spending less relative to their balance ends up with substantially more registered wealth surviving to the terminal-tax comparison, and the potential advantage of genuine non-residency scales up accordingly.

A retiree spending more has less left to arbitrage.

What this really describes is two different retirement objectives.

A consumption-focused retiree, someone who genuinely expects to spend most of their registered balance over their lifetime, is mainly working on an annual-withdrawal problem. Canadian marginal rates, treaty rates, destination-country tax, healthcare and lifestyle all matter, but terminal RRIF tax becomes progressively less important because there is progressively less RRIF left to tax.

An estate-focused retiree, someone whose registered assets materially exceed what they expect to consume, is working on a different problem. For that person, a large RRIF may genuinely survive to the second death. The resident-versus-non-resident terminal-tax gap can then become one of the largest Canadian tax variables in the plan.

Neither objective is inherently better. They simply produce different answers.

The couple version makes the same point even more sharply because couples add pension splitting and a spousal rollover.

For a couple with a combined CAD 2 million in RRSPs, split equally, modelled through a first death at 85 and second death at 90, the model produced roughly a CAD 357,000 advantage for genuine emigration at 70 when annual household spending was CAD 50,000.

At CAD 70,000 of spending, the modelled advantage fell to roughly CAD 244,000.

At CAD 100,000 of spending, it disappeared and became approximately a CAD 19,000 disadvantage relative to remaining resident.

All of those figures are before destination-country tax and specific to the assumptions of the model.

Even so, they make the important point:

“We have a CAD 2 million combined RRSP” is not enough information to answer the snowbird-versus-relocation question.

What the couple actually expects to spend can change the answer by hundreds of thousands of dollars.

There Is a Nasty Risk in Waiting Until Eighty

A minimum-withdrawal strategy built around emigrating at 80 can eventually benefit from non-resident taxation if the retiree actually gets there.

But the broader modelling found that taking minimums only was generally inferior to making moderate withdrawals before the move.

More importantly, this kind of plan carries a risk that no hindsight model can eliminate:

the retiree may die before becoming a non-resident at all.

A separate test built specifically to check this compared a plan of minimum withdrawals to age 79, with emigration planned at 80, against a straightforward resident strategy withdrawing a fixed CAD 70,000 a year, using the same CAD 1.5 million starting RRSP.

If death arrives before the planned move, while the retiree is still Canadian-resident and holding a RRIF deliberately left to compound largely untouched, the minimums-and-wait strategy comes out considerably worse than the steady resident strategy in this test.

Dying at 70 instead of reaching the planned emigration left the modelled estate roughly CAD 461,000 worse off than the fixed resident strategy.

Dying at 75 left it roughly CAD 543,000 worse off.

Dying at 79, one year short of the planned move, left it roughly CAD 597,000 worse off.

Surviving to the planned departure narrowed that gap sharply because the remaining RRIF finally became subject to the model’s non-resident treatment instead of resident terminal taxation. The shortfall fell to roughly CAD 83,000 at 80, CAD 107,000 at 85, CAD 113,000 at 90 and CAD 119,000 at 95.

Even so, in this particular test the minimums-and-wait strategy never caught the steady resident strategy at any age tested.

That is consistent with the broader pattern in this piece: the stronger later-emigration results came from moderate withdrawals before departure, not simply taking minimums and preserving the largest possible RRIF.

The mechanism is straightforward once stated plainly.

A wait-and-preserve strategy allows a large RRIF balance to grow while remaining fully exposed to resident terminal taxation for as long as the retiree remains Canadian-resident. If the planned departure never happens, all of that registered wealth remains inside the resident terminal-tax calculation the strategy was attempting to avoid.

None of this makes aggressive meltdown a universal rule either.

It means any strategy built around a future emigration date contains mortality and timing risk that a model with perfect foresight does not naturally capture.

The Country You Move To Can Destroy the Canadian Tax Saving

Everything discussed so far describes what Canada does.

It says nothing about what the destination country does.

That gap is large enough to erase much of the Canadian advantage in some cases and preserve much of it in others.

A 25 percent Canadian rate on a terminal RRIF amount is not necessarily a 25 percent total tax burden if the destination country also taxes the income, the estate, the heirs, or some combination.

The same caution applies to a 15 percent Canadian treaty rate on periodic RRIF withdrawals.

That is Canada’s rate.

It is not necessarily your final rate.

United States

The United States is a useful example.

The Canada-US treaty gives qualifying periodic Canadian pension payments, including RRIF payments within the applicable definition, favourable Canadian treaty treatment. The IRS also recognizes RRSPs and RRIFs as Canadian retirement arrangements and generally allows qualifying US taxpayers to defer US taxation on income accruing inside the plans until distribution.

But a US resident still has US federal income-tax obligations on distributions.

For Florida specifically, there is no state individual income tax, which makes the state materially different from moving to a higher-tax US jurisdiction.

On the estate side, the IRS’s 2026 basic federal estate-tax exclusion is USD 15 million per person, which keeps most ordinary Canadian retiree estates below the federal estate-tax threshold. That does not settle every cross-border estate or beneficiary-income issue, particularly for Canadians who become US persons, which is precisely why the US case needs to be modelled as a cross-border system rather than simply “15 percent RRIF tax.”

Portugal and Spain

Portugal and Spain are very different propositions.

Both generally tax residents on worldwide income, meaning the Canadian withholding rate is only one layer of the calculation.

Portugal is especially worth flagging because much of the expat content still circulating online is obsolete. Its old Non-Habitual Resident regime closed to most new entrants, and the replacement IFICI regime is not a general foreign-pension tax holiday.

I work through the current rules in Living in Portugal as a Canadian.

Spain likewise needs to be evaluated as a Spanish-resident tax problem, not simply a Canadian treaty-rate problem. Wealth taxation can also become relevant depending on the resident’s assets and region. The broader relocation case is covered in Living in Spain as a Canadian.

Italy and Greece

Italy and Greece complicate the picture in another direction because both have special tax regimes that can be attractive to qualifying foreign retirees.

Those regimes are real, but they have eligibility conditions and geographic or income-characterization rules that make “move there and pay 7 percent” far too simplistic.

The practical cases are covered separately in Living in Italy as a Canadian and Living in Greece as a Canadian.

Costa Rica and Panama

Costa Rica and Panama are commonly attractive to retirees because their tax systems are territorial rather than ordinary worldwide-income systems.

That can preserve more of the Canadian-side advantage where foreign retirement income remains foreign-source under the destination country’s rules.

But neither country has a Canadian income-tax treaty providing the kind of periodic-pension rate reduction available in countries such as the United States. Canada’s domestic non-resident withholding rules therefore remain important, and local sourcing and characterization still need to be confirmed rather than assumed.

Japan

Japan illustrates the opposite problem.

It is not a particularly compelling destination for someone whose main objective is Canadian retirement-tax arbitrage. Its tax treaty with Canada does not provide the familiar 15 percent private-pension framework available under the Canada-US treaty, and Japanese taxation becomes increasingly important as a Canadian establishes genuine residence there.

Japan also has a progressive inheritance-tax system with rates that can reach 55 percent at the top end. That does notmean a Canadian RRIF is automatically taxed at 55 percent. Liability depends on residence history, heirs, exemptions, the size and composition of the estate and other Japanese rules.

But it means Japanese inheritance taxation can materially reduce or potentially eliminate a Canadian terminal-tax advantage in the right circumstances.

For the much broader lifestyle, immigration and retirement case, see Living in Japan as a Canadian.

The consistent lesson across all of these destinations is not that one is universally better than another.

It is that the calculation has to be:

Canadian tax + destination-country tax.

Never Canadian withholding considered on its own.

Tax Is Only Half of the Relocation Decision

Even a thoroughly researched tax comparison answers only part of the question.

Healthcare may be the most consequential non-tax factor.

A Canadian who maintains provincial residence keeps provincial coverage, subject to the rules discussed earlier, although that coverage was never designed to finance serious healthcare received indefinitely abroad. Private travel insurance remains necessary for meaningful periods outside Canada.

A retiree who becomes a genuine non-resident needs a sustainable healthcare solution in the destination country instead, whether that means enrolment in its public system, private domestic insurance, international insurance or some combination.

Age and pre-existing conditions can make private coverage harder and more expensive over time.

That matters particularly for someone contemplating a tax-driven move at 75 or 80. The point at which the Canadian estate-tax arithmetic becomes most attractive may also be the point at which healthcare, mobility and family support make genuine relocation least attractive.

Immigration deserves equal weight.

The United States has no straightforward passive-income retirement visa for an ordinary Canadian retiree.

Mexico has temporary and permanent residence routes tied to financial qualification.

Portugal, Spain, Italy and Greece have residence pathways that can accommodate retirees or financially independent applicants, subject to country-specific requirements.

Costa Rica and Panama have established retiree-oriented residence programs.

Thailand offers retirement-oriented routes for qualifying applicants over 50.

Japan, notably, has no conventional retirement visa. Its Ministry of Foreign Affairs does offer a long-stay designated-activities status for qualifying nationals of visa-waiver countries with more than JPY 30 million in savings, but the initial stay is six months, extendable to a maximum of one year, and dependent children cannot accompany the applicant under that program.

That is a very different proposition from conventional retirement residence.

Europe has another constraint. A Canadian who has not obtained residence status generally remains subject to the Schengen Area’s 90-days-in-any-180-days visitor rule. The days are counted across the Schengen Area, not separately for Portugal, Spain, Italy, Greece or France.

This is one reason a classic five- or six-month Canadian snowbird lifestyle works much more naturally in some destinations than others.

And finally, family and actual lifestyle preference matter in a way that is easy to lose under a spreadsheet full of tax rates.

A plan that saves CAD 200,000 of lifetime Canadian tax but requires living somewhere you do not actually want to live, away from children, grandchildren, friends and a familiar support system, is not automatically a better retirement plan.

Tax is an input to the decision.

It is not the decision.

Three Different Strategies That Often Get Confused

Much of the confusion around this topic comes from treating “leaving Canada” as a single decision.

In practice, it describes at least three very different strategies.

1. Snowbird Strategy

The snowbird keeps Canadian tax residence while spending substantial time abroad.

This suits someone who wants Canada to remain the base of operations, values provincial healthcare and family proximity, and wants a warm-weather season rather than a permanent move.

The tax treatment remains ordinary Canadian resident taxation.

The planning challenge is primarily staying inside Canadian provincial-health requirements and the immigration and tax-residence rules of wherever you spend the winter.

2. Lifestyle Relocation

The lifestyle emigrant moves because they genuinely want to live somewhere else.

Tax still matters, but it is not the reason for the move.

For this person, the correct order of operations is usually to decide where they actually want to live and then understand what that country’s tax, healthcare and immigration system does to their finances.

This is the question behind the broader Sovereign Canadian Expat Living series: not where the lowest tax rate appears on a spreadsheet, but what actually living somewhere else buys you and costs you.

3. Tax and Estate-Driven Relocation

This is the narrowest group and the one this article has spent the most time investigating.

It describes someone with more registered assets than they expect to consume, a reasonable expectation of leaving a substantial RRIF at death, genuine willingness to sever Canadian residential ties, and a destination whose own tax system does not quietly erase the Canadian-side advantage.

This is the group for whom detailed personalized cross-border modelling can genuinely be worth the professional fees.

The amounts involved can run into hundreds of thousands of dollars.

But the move has to be real.

A tax strategy that only works if you genuinely cease Canadian residence cannot be implemented by calling yourself a non-resident while continuing to live an essentially Canadian life.

So Where Are the Rough Thresholds?

Useful ranges, not magic numbers.

Spending, investment returns, marital status, income distribution between spouses, destination taxation and longevity all move them.

For a single retiree with moderate registered assets, around CAD 500,000 in the scenarios modelled here, staying Canadian-resident and drawing the account down deliberately tended to remain competitive or superior.

Around CAD 1 million, the answer became substantially more sensitive to spending, returns and relocation age.

From roughly CAD 1.5 million upward, genuine relocation became materially more interesting in the model, particularly where the retiree expected substantial registered wealth to survive to death rather than be consumed.

Couples are different.

Equal registered balances combined with pension income splitting make Canadian residency more competitive than the single-retiree figures might suggest at the same household balance.

At roughly CAD 1.5 million combined, the relocation advantage in the scenarios examined was generally modest and highly sensitive to assumptions.

Around CAD 2 million combined, it became more material at moderate spending levels.

But in the fixed-spending example discussed earlier, increasing annual household spending to CAD 100,000 was enough to erase it.

Which brings us back to the genuinely useful number:

How much RRIF are you realistically likely to be holding at the second death?

That number may matter more than today’s RRSP balance.

A Simple Decision Framework

Before assuming either strategy is obvious, work through these questions with actual numbers:

1. Do you actually want to live abroad, or are you mainly trying to escape Canadian winters for part of the year?
2. How much RRSP or RRIF are you realistically likely to be holding around age 70?
3. How much do you actually expect to spend each year in retirement?
4. Given that spending, how much RRIF is realistically likely to remain at the second death?
5. What Canadian marginal tax would relocation actually let you avoid?
6. What treaty rate, if any, applies to regular withdrawals in your intended destination, and does it apply only to periodic payments?
7. What will the destination country itself tax during your lifetime?
8. What will that country tax at death or when your heirs receive the assets?
9. What happens to your OAS given your actual Canadian residence history?
10. How much pension-splitting value would you give up?
11. Can you legally remain in the destination for the amount of time your plan requires?
12. Can you maintain adequate healthcare there into your eighties?
13. What happens financially if you die before the planned move ever happens?

None of these questions has a universally correct answer.

Working through them honestly does more useful planning work than comparing a Canadian 53.5 percent top marginal rate with a foreign 15 or 25 percent withholding rate and declaring the case closed.

Conclusion

The question this piece opened with, framed as a straightforward choice between snowbirding and leaving, was never going to resolve into a single clean answer.

For a retiree with a moderate RRSP who genuinely plans to spend most of it, the familiar Canadian approach remains difficult to dismiss on tax grounds alone: deliberately draw registered assets during lower-income retirement years where appropriate, retain Canadian residence and the benefits that come with it, and spend part of the year elsewhere within provincial-health and foreign-immigration limits.

The modelling in this piece does not support abandoning that approach simply because another country offers a lower Canadian withholding rate.

For someone holding substantially more registered wealth than they expect to consume, however, the calculation changes.

At that point, the treaty rate applying to annual withdrawals matters.

The pension splitting given up by leaving matters.

The OAS consequences of an early departure matter.

Destination-country taxation matters enormously.

And if a seven-figure RRIF is realistically going to survive to the second death, the Canadian tax treatment of that remaining account can become one of the largest variables in the entire retirement plan.

Put plainly:

For someone likely to consume most of the RRIF, annual retirement taxation is the dominant tax problem.

For someone likely to reach the second death with a seven-figure RRIF still intact, residence at death becomes a major estate-planning variable as well.

That does not mean an 80-year-old should move abroad to save tax.

It means a Canadian who genuinely wants an international retirement and is likely to leave a very large registered estate should understand that the tax consequences of residence can be much larger than the conventional snowbird discussion suggests.

The most useful finding to come out of this research, more than any single tax rate or threshold, is therefore a different question entirely:

Not snowbird or emigrate, but how much registered wealth will you actually still have when you die, and where will you genuinely be resident when that happens?

Everything else in this piece is really an attempt to answer that question honestly.

This article is general information for Canadians researching retirement, tax residency and life abroad. It is not individualized tax, legal, investment, estate-planning, healthcare or immigration advice. Cross-border tax treaties and foreign tax regimes can materially change the result, and the modelling above uses simplified assumptions rather than predicting any reader’s actual outcome. Anyone contemplating a change of tax residence, particularly with a substantial RRSP or RRIF, should verify the current rules with qualified Canadian and destination-country professionals before acting.