Insured Retirement Plan

An insured retirement plan is a leveraged strategy.

A participating policy is funded heavily for many years. In retirement, it is assigned to a lender as security for a line of credit, and the borrowing provides income. At death, the death benefit repays the lender and whatever remains goes to the estate.

A loan advance is not income, so it is not taxed on receipt. That single fact is the entire appeal, and everything that can go wrong follows from what has to be true for it to keep working.

How it actually works

Fund a participating policy heavily, typically through a paid-up additions rider, within the limits the exempt test allows under Regulation 306, Income Tax Regulations.

Let it accumulate, usually for fifteen to twenty-five years. Contractual value builds on the guaranteed schedule, and dividends may add to it.

At retirement, assign the policy to a lender as collateral for a line of credit.

Draw against the line, annually or monthly, as retirement income.

Interest accrues and is usually capitalised rather than paid, so the balance grows each year.

At death, the death benefit repays the lender first. Any remainder goes to the named beneficiary or the estate.

Why the loan comes from a lender and not the insurer

This is the technical point that decides whether the structure works, and it is routinely stated wrongly.

An advance from the insurer is a disposition. Under ITA s.148(9), amounts above the adjusted cost basis are taxable. Taking retirement income that way would produce a tax bill every year, which is precisely what the structure exists to avoid.

A loan from a third-party lender is not a disposition. The policy is assigned as security, ownership does not change, and no disposition occurs. The advance is borrowed money.

So the arrangement requires an outside lender, and that requirement is the structure's principal weakness rather than an administrative detail.

Any description of an insured retirement plan that says the income comes from policy loans has described a different and taxable arrangement. The earlier version of this page said exactly that.

What it depends on

Five conditions, and none is guaranteed.

That a lender will still lend. Lending against an assigned policy is a commercial decision, reviewed periodically. It is not a contractual entitlement and the appetite for this lending has changed before.

That the loan-to-value limit is not breached. Lenders advance a proportion of the contractual value. If dividends underperform and the balance grows faster than the value, the limit can be reached, at which point drawing stops and repayment may be demanded.

That interest rates remain tolerable. The balance compounds at whatever rate applies. A period of high rates accelerates the balance against a value that grows on its own schedule.

That dividends broadly hold. Projections are built on a scale that is declared annually at the discretion of the insurer's board and is not guaranteed. It has moved historically and can move again.

That the tax treatment is unchanged in thirty years. The current treatment is established, and a strategy relying on a specific treatment across three decades carries the risk that it changes.

How it fails

Worth setting out plainly, because the failure is not gradual.

The facility is reduced or withdrawn. Income stops, and the accumulated balance remains.

The loan-to-value limit is reached. Drawing stops and repayment may be required.

The balance is repaid by surrendering the policy. The contract ends, the coverage is gone, and the accumulated gain above the adjusted cost basis becomes taxable, frequently in a single year, at an age when income is otherwise low but the amount is large.

And the adjusted cost basis is usually near nil by then, because it declines over the life of a long-held contract, so almost the entire value is taxable.

The person affected is in their seventies or eighties, with no coverage, a tax bill, and no time to rebuild. That is the scenario the structure has to be evaluated against, not the illustration.

The costs

The insurance costs, which the earlier version of this page disclosed honestly and which are worth repeating: mortality charges, administration, premium tax and acquisition cost, weighted heavily to the early years.

The lending costs: setup, annual review fees, and interest that compounds against the value securing it.

And the opportunity cost of heavy funding sustained for two decades, measured against what that money would otherwise have done. That comparison is the honest test and it is set out with the money principles this site works from.

This page does not publish projected growth rates or cost percentages. The earlier version stated that value "typically grows at 4 to 6% annually", which is a performance projection presented as typical without a source or an insurer attached. Ask for the guaranteed column on an illustration prepared for you.

Against an RRSP

An RRSP gives a deduction now and taxes the withdrawal fully as income later. It is the more efficient instrument for most Canadian households and it should be used first.

A TFSA gives no deduction and is genuinely tax-free on withdrawal, with full liquidity and no leverage.

This structure gives neither deduction nor certainty, and requires substantially more capital, a much longer horizon and a lender's continued cooperation.

It becomes worth examining only after registered room is fully used, and for a household with surplus capital, a genuine permanent insurance need, and the capacity to absorb the failure case. For an incorporated physician the corporate balance and the passive income position come first.

Who this does not suit

Anyone with unused registered room.

Anyone whose retirement would be in difficulty if the facility were withdrawn. If the income is essential rather than supplementary, the dependency is too great.

Anyone with a horizon under fifteen years.

Anyone who cannot sustain heavy funding through a poor decade, because a plan built on a good year fails in a normal one.

Anyone who does not want permanent insurance in its own right. The coverage should be wanted independently. Where it is not, the structure is a financing arrangement wearing an insurance policy.

How it differs from the approach Nelson Nash named

Related and materially different, and the distinction is worth keeping.

The Infinite Banking Concept® is a term originated by Nelson Nash and a registered trademark of Infinite Banking Concepts, LLC. Neither this practice nor its author is affiliated with, sponsored by or endorsed by that company or the Nelson Nash Institute. Neither this practice nor any policy is a bank.

That approach generally uses advances from the insurer during working life, repaid deliberately, and does not depend on an outside lender.

This structure depends on an outside lender entirely, in retirement, when there is no earned income to fall back on.

The second carries a risk the first does not. Anyone who has read about the first should not assume the second inherits its characteristics.

What stands behind any of this

The guarantees are the insurer's and depend on its solvency. They are not backed by any government. Assuris protects Canadian policyholders within published limits.

Dividends are declared annually at the discretion of the insurer's board and are not guaranteed.

The lending arrangement is guaranteed by nobody.

The questions to put to anyone proposing this

Nine, and the answers should be in writing.

Show me the guaranteed column. What the contract does if no dividend is ever paid, alongside cumulative premiums, at years five, ten, twenty and at retirement.

Model it with the dividend scale one full point lower. If that cannot be produced, you have been shown a single scenario and told it is a plan.

Model it with interest rates three points higher than the illustration assumes, sustained for a decade.

Which lender, and on what terms? Name the institution. Show the loan-to-value limit, the review frequency, and what happens if the limit is breached.

Is the lending commitment contractual for the life of the arrangement? The answer is almost always no, and hearing it said aloud is the point.

What is the tax bill if this unwinds in year twenty-five? A number, prepared by an accountant, not a reassurance.

What is the projected adjusted cost basis at retirement? If it is near nil, almost the entire value is taxable on a surrender, and that should be stated rather than discovered.

Who services this in twenty years? The arrangement outlives most advisory relationships and requires active management throughout.

What is your compensation on this, and how does it compare to funding my registered room instead? A fair question, and the reaction to it is informative.

What the illustration does not show

An illustration for this structure is a projection under assumptions held constant for decades. Several things it omits are the things most likely to matter.

That the lender is a party with its own interests, reviewing the file periodically against its own risk appetite and the regulatory environment it operates in. None of that appears in a spreadsheet.

That capitalised interest compounds against a value growing on a different schedule. Two curves, and only one of them is contractual.

That the failure case is not a lower number. Illustrations show outcomes varying by degree. This structure has a discrete failure mode: the facility ends, the policy is surrendered, and a large gain crystallises in one year.

That the person managing it will be older. The arrangement requires attention in the decades when people are least equipped to give it, and cognitive decline is a real risk to anything requiring active management into advanced age.

Ask for the failure case as its own page, with its own numbers. A proposal that cannot produce one has not been thought through.

Where it does make sense

Stated fairly, because a page that only sets out risks has not described the thing.

A household with a genuine permanent insurance need, wanted for its own sake, independent of any income strategy.

Registered room fully used, year after year, with surplus remaining.

Capital sufficient that the arrangement is supplementary, not the retirement plan itself. A property investor whose capital is already illiquid rarely meets that condition.

A horizon of twenty years or more before drawing begins.

Advisors who have seen one unwind, and an accountant who has modelled the failure case in writing.

Where all five are true, the structure is defensible and the tax treatment of borrowed money is real rather than a trick. Where any one is absent, the case weakens sharply, and where two or more are absent it should not proceed.

That is a narrow set of circumstances, and stating how narrow is the useful part. This arrangement is presented to far more households than it fits, because the illustration is persuasive and the failure case is thirty years away and appears on no page of it.

A reader who works through the five conditions and finds two missing has learned something worth more than any projection: the arrangement is not for them, and they now know why rather than having been told.

And a reader who finds all five true still has work to do. The next step is not a signature. It is an accountant producing the failure-case number in writing, and a lender naming its terms, before anything is funded.

The names this arrangement travels under

Worth listing, because the same structure is presented under several labels and a reader may not recognise it as the thing described here.

Insured retirement plan, or IRP. The most common term.

Insured retirement program, used interchangeably.

Leveraged insured retirement plan, which is the most accurate of them, since it names the borrowing.

Cash flow plan, or variations that avoid the word retirement.

Corporate insured retirement plan, where a company owns the policy and the borrowing arrangement sits at the corporate level. The mechanics are the same and the tax analysis is different, involving the shareholder benefit rules and the Capital Dividend Account.

And sometimes no name at all, presented simply as a way to use a policy for retirement income.

If borrowing against an assigned policy is involved, it is this arrangement, whatever it is called, and everything on this page applies.

The three parties, and whose interests align with yours

Setting them out separately makes the structure easier to evaluate than any diagram does.

You. You want income, coverage that persists, and no forced unwinding.

The insurer. It has issued a contract with a guaranteed schedule and it administers the policy. Its obligations to you are contractual and do not change because you assigned the policy. The insurer is not party to the lending arrangement and has no duty to keep it available.

The lender. It advances money against security, reviews the file periodically, and manages its own risk. It has no obligation to continue lending, and its interests diverge from yours precisely when conditions deteriorate: if the value supporting the loan weakens, the lender's correct response is to reduce exposure, which is the moment you most need it not to.

That divergence is the structural weakness. Nothing improper is happening when a lender withdraws. It is doing what a lender does.

And the advisor is a fourth party whose compensation arrives when the policy is issued, years before any of the above is tested.

What to do if you already have one

A different question from whether to start one, and the more urgent for anyone partway through.

Find out where you stand. The current contractual value, the outstanding loan balance, the loan-to-value limit, and the headroom between them. Four numbers, available from the insurer and the lender.

Establish the adjusted cost basis. This determines the tax bill if the structure ever unwinds, and it is the number nobody volunteers.

Ask when the facility was last reviewed and when it will be next.

Model the unwind now, while there is time. An accountant can calculate what a surrender would cost today. Knowing that number is better than discovering it.

Consider whether to stop drawing. Where the headroom is thin, reducing or pausing withdrawals extends the arrangement's life considerably, and it is easier to reduce spending deliberately than to have the facility reduced for you.

Do not surrender in reaction. A surrender crystallises the entire gain in one year. Where an exit is genuinely necessary, spreading it or exploring alternatives with an accountant first is almost always better than acting quickly.

Get a second opinion from someone who did not arrange it. This is the single most useful step, and the reluctance people feel about it is worth overriding.

What this page will not do

It will not recommend the structure.

It is among the most complex arrangements sold in Canadian insurance, it depends on conditions that persist for decades and are outside anyone's control, and it fails badly rather than gently. Where it is appropriate it should be arranged by people who have seen one unwind, with an accountant modelling the failure case in writing before anything is signed.

Everything here is written by someone paid by commission from an insurer when a contract is issued, which is stated on the author page and at the foot of every page.

The wider retirement context is in retirement planning, and the mechanics of advances against a contract are in how a participating policy works, year by year.