What Critics Get Right
Several standard criticisms are correct. Five are conceded as stated: the wrong comparison, interest paid to the insurer, no fund-style cost figure, costs weighted to the early years, and sales to people it does not suit. Three are conceded in part, with the limit named. The ninth, on tax, holds under Canadian law and misfires only when it borrows American rules. A participating policy is insurance, not an investment, and a policy loan is an advance from the insurer.
Several of the standard criticisms of this approach are correct, and they are worth reading at their strongest before anyone shows you an illustration. Of the nine set out below, five are conceded as stated. Three are conceded in part, with the limit named. The ninth, on tax, is correct under Canadian law and misfires only when it borrows American rules. None of that makes the product unlawful or useless. It makes it narrow, and it leaves the decision with your household's figures.
The approach is the financing approach known as The Infinite Banking Concept®, set out by R. Nelson Nash: fund a specially designed, high-cash-value, participating whole life insurance policy, let the cash value build, and use policy loans for large purchases, repaying them on a schedule the owner sets. The product underneath is insurance. It is not a savings account or an investment, and no deposit insurance applies to it. The wider case against, and what it gets right, is set out in the objections and risks section.
How does the arrangement work before anyone criticises it?
Every criticism below turns on four roles, so it helps to name them. The owner holds the contract and makes its decisions. The person insured is the life the contract covers. The owner and the person insured can be the same person, or two different people. The beneficiary receives the death benefit when the person insured dies. The insurer issues the contract, pays that benefit, and is the party the owner borrows from.
A policy loan is an advance from the insurer, made under the contract and secured by the policy's cash value. The contract sets how much can be advanced. The insurer sets the interest rate and may change it as the contract allows. You owe the insurer, and the interest is paid to the insurer. Depending on the contract, unpaid interest can be added to the balance. While a loan is outstanding, the insurer deducts the balance and the accrued interest from what is paid when the person insured dies. If the balance grows past the value securing it, the contract can end under its terms.
Two more points sit inside the contract. Some contracts change the dividend credited on the borrowed portion of the value, and some do not; ask where yours says so. And the dividend scale interest rate an insurer publishes is one input to the dividends it declares. It is not the return on your policy.
There are three ways to pay for a purchase, and they differ in who lends and who is owed:
| Route | Who advances the money | Who receives the interest | Who sets the rate | At the death of the person insured |
|---|---|---|---|---|
| Policy loan | The insurer, under the contract | The insurer | The insurer, which may change it | Balance and interest are deducted from the death benefit |
| Loan from an outside lender, with the policy assigned as security | The lender | The lender | The lender, under its own terms | The lender, as assignee, can be paid from the death benefit |
| Paying from savings | Nobody | Nobody | Not applicable | No effect on the policy |
The AMF describes the same mechanics for Quebec readers: a policy advance is a loan that bears interest, and you can instead borrow from another institution with the insurance as security (AMF, in French). The details are at how policy loans work.
What is the verdict on each criticism, at a glance?
| No. | The criticism | Verdict | What settles it for you |
|---|---|---|---|
| 1 | The usual comparison is against the wrong alternative | Conceded | What you would really have done with the purchase |
| 2 | Interest on a policy loan does not come back to the owner | Conceded | The loan provision: who lends and who is paid |
| 3 | Costs are not disclosed the way a fund's are | Conceded as stated | The guaranteed schedule, read year by year |
| 4 | Costs fall early, and the buyer carries the cost of leaving | Conceded | Cumulative premiums beside the guaranteed surrender value |
| 5 | Policies are sold to people they do not suit | Conceded | The advisor's documented suitability assessment |
| 6 | Registered room should always come first | Conceded in part | A professional licensed for registered plans, with your figures |
| 7 | The vocabulary overstates what the owner controls | Conceded in part | The contract's loan provision |
| 8 | Illustrations are treated as forecasts | Conceded in part | Current and reduced-scale illustrations beside the guaranteed column |
| 9 | The tax advantages are overstated | Correct under Canadian law; misdirected when it relies on American rules | The adjusted cost basis and an accountant |
Where a criticism is conceded, no rebuttal follows it, because a concession with a rebuttal attached is not a concession.
1. Is the usual comparison made against the wrong alternative?
The argument. Supporters compare a policy loan with a loan from an outside lender and conclude that the owner keeps interest that would otherwise have left the household. But some purchases were never going to be financed with a loan. For those, the realistic alternative is paying from savings, which costs no borrowing interest, although it uses your liquidity and gives up what that money would otherwise have earned.
Verdict: conceded. It is correct, and it goes to the heart of the usual sales case. Measured against paying cash, the advantage shrinks and can disappear. A presentation that does not name the alternative it is measuring against has left out the variable that decides the result. The argument is worked through at the comparison question.
2. Does the interest on a policy loan come back to the owner?
The argument. Presentations imply that interest paid on a policy loan returns to the owner. It does not. It is paid to the insurer.
Verdict: conceded. The insurer advances the money, the owner owes the insurer, and the interest belongs to the insurer. Any presentation that implies dollar-for-dollar recovery is describing something that does not happen. A repaid loan restores the value available to borrow against; it does not return the interest.
3. Are the costs disclosed the way a fund's are?
The argument. A fund publishes a management expense ratio: one number, standardised and comparable. A participating policy publishes nothing equivalent. The cost of insurance, the contract's charges and the advisor's compensation sit inside the contract and are not itemised for the buyer as one figure.
Verdict: conceded as stated. There is no single fund-style figure. The stronger word some critics use, hidden, goes further than the facts: the guaranteed schedule prices the whole structure in numbers you can read, and the contract lists the charges it states. That makes the schedule a good tool for knowing what you own and a poor one for comparing products. How to measure the cost anyway is at the real costs.
4. Does the buyer carry the cost of leaving early?
The argument. Compensation and the cost of putting a policy in force fall heaviest in the first years, which is when a buyer is most exposed to a change of mind or a change of income. The structure puts the cost of an early exit on the person least equipped to price it at the signing table.
Verdict: conceded. The figure is in the guaranteed column from the start. Put cumulative premiums beside the net guaranteed surrender value at years 1, 3, 5 and 10. The difference is the cost of leaving at that point; it is not a separately disclosed fee.
Illustrative example. Assume you pay $8,000 a year for five years, $40,000 in total, and the guaranteed cash surrender value at the end of year five is $29,000. The figures are chosen to show the arithmetic and come from no contract. Surrender then and you receive $29,000, less any loan outstanding, and the $11,000 difference is gone. Your own illustration shows the real figures.
5. Are policies sold to people they do not suit?
The argument. Compensation is paid when a policy is issued, while suitability depends on funding sustained for decades. Those two facts create pressure in one direction.
Verdict: conceded. The pressure is real, and some policies are sold to people they do not suit. The same compensation structure exists across advised insurance products in Canada, which describes the market rather than excusing it. Assessing suitability is the advisor's obligation, and it should be documented in writing before any recommendation, for example in a needs analysis. The questions further down let you check that it was done. How the firm behind this site is paid is stated on the author page, and the households most exposed are named at risks and failure modes.
6. Should registered room always come first?
name the alternative, or there is none
The comparison that is actually honest
- The usual case compares an advance to an outside loan
- That holds only if you would have borrowed anyway
- If you would not have, compare it against paying cash
- Interest on an advance is paid to the insurer
- A comparison is incomplete until the alternative is named
The argument. Unused TFSA or RRSP room should always be filled before a permanent policy is considered as a way to build savings.
Verdict: conceded in part. What holds: premiums and contributions can draw on the same surplus dollars, so a policy can displace registered room, and a presentation that never asks whether it would is incomplete. Selling permanent insurance for accumulation while ignoring a household's other options raises a serious suitability concern. Whether a particular recommendation was improper depends on the facts and on the provincial rules that govern the advisor.
What does not hold is the word always. A TFSA, an RRSP or an FHSA does a different job from life insurance. This practice is licensed to place insurance of persons and is not registered to advise on registered plans, so it gives no ordering between them and a policy, in either direction. A general rule on that ordering, whichever way it points, is a sales argument rather than advice. Put the registered-plan side of the question to a professional licensed for it, and the tax side to an accountant, with your own figures in hand.
7. Does the vocabulary overstate what the owner controls?
The argument, in the critics' own words. The name The Infinite Banking Concept® and the title of Nelson Nash's book, Becoming Your Own Banker®, are marks of Infinite Banking Concepts, LLC. Critics say they promise something a contract cannot deliver: an institution of your own, and independence from lenders.
Verdict: conceded in part. The criticism of the language is correct. The words name a body of literature, not anything a Canadian licensed advisor provides. What the owner holds is a contract, administered by the insurer under its terms. Nobody becomes an institution.
The limit is that the feature underneath is real. The contract gives the owner a right to request an advance from the insurer against the cash value, up to a limit the contract sets, without applying to an outside lender. The insurer sets the rate and may change it, and the contract, not the owner, governs the conditions. That is a genuine difference in access. It is not independence, and it should not be described as if it were.
8. Are illustrations treated as forecasts?
The argument. A document showing values decades out is arithmetic under assumptions. Dividend scales move. Presenting a projection as an expectation misleads, whether or not anyone intends it.
Verdict: conceded in part. The misuse is real. The limit is that an illustration also carries a guaranteed column, and that column is contractual. It is not unconditional, though. The guaranteed values assume every required premium is paid and nothing is borrowed or withdrawn beyond what the illustration shows. They are the insurer's obligation, so they depend on the insurer standing behind them. Values that depend on future dividends, including paid-up additions not yet purchased, are not guaranteed.
If a member insurer fails, Assuris states that a whole life policyholder keeps up to $1,000,000 or 90% of the death benefit, whichever is higher, and up to $100,000 or 90% of the cash value, whichever is higher, both calculated after deducting policy loans (Assuris, whole life).
The practical answer is to ask for three things side by side: the guaranteed column, the illustration at the current dividend scale, and an illustration at a reduced scale. The gap between them measures how much of the picture is assumption. Once the policy is issued, check the guaranteed figures against the schedule in the contract itself.
9. Are the tax advantages overstated?
the commonest reasons it fails
Who this method does not suit
- 01A household whose income cannot carry an ordinary decade
- 02Anyone who may need the capital in the first several years
- 03Anyone who will not repay what they draw
- 04Anyone who does not actually want permanent coverage
- 05Anyone who cannot say what the contract is for
The argument. The tax treatment of a policy is narrower than sales presentations suggest. Some versions of this argument reason from American rules: the modified endowment contract rules or section 7702 of the US Internal Revenue Code.
Verdict: correct under Canadian law; misdirected when it relies on American rules. The American rules do not govern a Canadian contract. The Canadian rules do, and they carry their own cautions, which are the ones a Canadian reader needs:
- Growth inside a policy is not taxed each year while the policy remains exempt under Regulation 306 of the Income Tax Regulations. A policy that ceases to be exempt is deemed to have been disposed of (paragraph 148(2)(d) of the Income Tax Act).
- A policy loan is a disposition (subsection 148(9), definition of "disposition", paragraph (b)). Only the part of the loan proceeds above the adjusted cost basis immediately before the loan is income (subsection 148(1)).
- The loan lowers the adjusted cost basis, so a later loan can find less basis left. Repaying restores it and can give a deduction in the year of repayment, up to amounts previously included in income (paragraph 60(s)).
- Assigning the policy as security for a loan from another lender is not a disposition (paragraph (f) of the same definition).
- A surrender, or a lapse with a loan outstanding, can produce income to the extent the proceeds exceed the adjusted cost basis. For tax, those proceeds are the cash surrender value less the policy loans owing, so a small cheque does not mean a small gain.
- Loan interest can be deductible only when the borrowed money is used to earn income from a business or property, and then only as the insurer verifies it on CRA Form T2210. Interest on money used for personal spending is not deductible.
- A death benefit paid under an exempt policy because of the death of the person insured is not a disposition (paragraph (j)).
Federal rules apply everywhere in Canada; Quebec residents also file with Revenu Québec. That is our reading of the provisions, not a ruling, and your accountant should apply it to your figures. How the exempt test works is at the exempt test, and the other ways American material misleads a Canadian reader are at American videos and Canadian law.
Is it a scam?
No. A specially designed, high-cash-value, participating whole life insurance policy is an insurance contract governed by provincial insurance law, including the Civil Code in Quebec. It is issued by an insurer whose solvency is supervised according to its charter: by OSFI for a federally incorporated insurer, and by its home province for a provincially incorporated one, which in Quebec means the AMF. It is sold by agents whose licences appear in public registers kept by the licensing bodies, such as the AMF, FSRA in Ontario and the Insurance Council of British Columbia. There is no recruitment structure and no return paid out of later buyers' money, and the terms are in a document you can read before signing.
The word scam describes fraud. Applying it to a regulated contract makes the real criticisms above easier to dismiss, which helps nobody. Lawful is still not the same as suitable. The longer answer, including how to check a licence, is at is it legitimate?
Does the insurer keep the cash value when the person insured dies?
Partly, in the sense critics mean. When the person insured dies, the beneficiary receives the death benefit, not the death benefit plus the cash value. The AMF says so plainly: the surrender value is not payable on death; the amount of insurance is. On that point the objection describes the payout correctly.
What it misreads is the structure. The cash surrender value is not a separate account sitting beside the contract; it is a value inside the same contract. As it grows, the insurer's amount at risk, the death benefit minus the cash value, shrinks. The cash value is part of what funds the death benefit, so nothing extra is taken at death, but the pure insurance protection does shrink over time. Dividends used to buy paid-up additions can raise both figures, depending on the dividend option chosen. Any policy loan and interest owing are deducted from what is paid.
Contracts differ at advanced ages, and no single pattern describes them all. Some are designed so that the guaranteed cash value reaches the face amount at a stated age; others keep the coverage in force past 100 on terms the contract sets. The maturity of a policy is a disposition for tax (subsection 148(9), definition of "disposition", paragraph (c)). Ask for the guaranteed cash value and death benefit rows at ages 85 and 100, and for the clause that says what happens at 100. Whether the whole is good value is a fair question, and the contract, not the slogan on either side, answers it.
Is buying term and investing the difference the better route?
Sometimes it is. When the need for coverage is temporary, such as a mortgage or children who will become independent, term insurance can do the insurance job for much less, and saying so is the honest position.
Stated as a universal rule, it assumes four things: that the difference is actually invested rather than spent, that it stays invested through a bad market, that the term coverage is renewed or replaced when it ends, at an older age and a higher price, and that you are still insurable then. Where the need for coverage is permanent, those assumptions carry the weight of the argument.
Both routes need discipline: investing the difference for one, paying premiums for decades for the other. A fair comparison puts both under the same assumption about whether you will keep going, prices the death benefit on both sides, and uses your own figures.
What do critics and practitioners both accept?
each one is wrong, and correctable
Claims that should never be made
- 01That you are borrowing your own money
- 02That you pay the interest to yourself
- 03That an advance leaves the contract untouched
- 04That it replaces a registered plan
- 05That the dividends are guaranteed
More than either side admits, and it is the ground a reader can stand on:
- The costs are front-loaded, and leaving early is expensive.
- Dividends are not guaranteed.
- A household without durable surplus has cheaper places for its money.
- Illustrations are projections, and they are misused when read as forecasts.
- The product needs a horizon measured in decades.
- Some policies are sold to people they do not suit.
None of the six depends on an opinion about the method. They describe the contract, how it is priced and how it is distributed. A reader who holds these six is well equipped to weigh anything either side says next.
What should an advisor tell you before you ask?
Some things are true of this product and belong in any presentation, whether or not the buyer raises them:
- The first-year numbers look poor. A policy's guaranteed cash value in its early years sits well below what has been paid. Attention directed only to year twenty skips the years in which a change of mind costs the most.
- A current-scale illustration alone is incomplete. Insurers can illustrate values at a lower dividend scale. Ask for one, because it shows how the picture changes if dividends fall.
- An early surrender leaves you with less than you paid. Ask for year-by-year surrender values and a stop-paying scenario at year four or five, so the cost of changing course is on paper.
- Compensation paid at issue shapes behaviour. Ask how the advisor is paid on this policy, and when.
None of that makes the product wrong. It makes distribution the weak point, which is what the fifth criticism already says.
What do the nine criticisms not establish?
A list of concessions can read like a verdict. It is not one.
None of the nine shows that the product does not work. Together they show that it is expensive as a way to grow money, unforgiving of an early exit, hard to compare on cost, and sold to some people it does not suit. A product can be all of that and still be right for a particular household, which is what suitability means.
None of them addresses the death benefit, which is the thing actually being bought: a sum paid when the person insured dies, whenever that happens, that no portfolio provides.
And none of them is an argument about you. Every criticism here is general. Whether any applies to your situation depends on facts that only your household and your contract hold, which is why the next section lists them.
What should you collect before you decide?
five components, each behaving differently
What a participating contract costs
- 01The mortality chargeBuys the death benefit.
- 02CompensationWeighted to the first year.
- 03Policy and administration feesBuilt into the premium; ask the insurer which are stated separately.
- 04Provincial premium taxIncluded in the premium.
- 05Loan interestOnly if capital is actually accessed.
Start with your own facts. The criticisms land hardest where one of these is missing:
- A real want for permanent coverage for its own sake, not only for the cash value.
- Surplus that survives an ordinary year, and a poor one, not only a good year.
- No need for the money within the first ten years.
- No high-rate debt outstanding.
- An emergency reserve outside the policy.
- Insurability: the underwriting result, since a rating changes the cost and a decline ends the plan. The outcomes are at rated, postponed or declined.
Then collect the contract's facts. Leave the figures to the insurer; what matters is that every row is on paper before you sign:
| Ask for | At which years or ages | What it answers |
|---|---|---|
| Guaranteed cash surrender value and guaranteed death benefit | Years 1, 3, 5, 10 and 20; ages 85 and 100 | Criticisms 3 and 4: the cost of leaving early, and what the contract promises without dividends |
| Values at the current dividend scale | The same rows | What the insurer projects if today's scale continued, which it may not |
| Values at a reduced dividend scale | The same rows | Criticism 8: how much depends on dividends |
| A stop-paying scenario | Stopping at year 4 or 5, and at year 10 | Whether coverage continues, at what amount, under which option |
| A loan scenario | A loan in a year you choose, repaid and not repaid | Criticism 2: how the loan, its interest and the death benefit interact |
| The adjusted cost basis | Today, and in the year of any planned loan | Criticism 9: how much of a loan or a surrender could be income |
| The loan provision and the clause on what happens at 100 | Not applicable | Criticism 7: the limit, how the rate is set, and whether the policy matures or continues |
Each question has someone who can answer it:
| Who to ask | What to ask |
|---|---|
| The advisor | Why the coverage is permanent; which premium survives a poor year; who should not buy this; how the advisor is paid, and when |
| The insurer, in writing | How the loan rate is set and whether it can change; whether a loan changes the dividend credited; which options apply if premiums stop; what it would report for tax on a loan or a surrender |
| An accountant | What a loan, a surrender or a lapse would mean on your return; whether any interest could be deductible; for a corporation, its capital dividend account |
| A professional licensed for registered plans | How the registered side of your plan fits, with your own figures |
| A lawyer, or in Quebec a lawyer or notary | Beneficiary designations, an assignment to a lender, and a written agreement for any family loan |
Which questions test a presentation against the nine?
Five questions cover the nine, and none needs technical knowledge:
- Which of the nine applies to what you are showing me? An honest answer names at least one without prompting. This covers criticism 1, and the follow-up is what you would compare against if you would have paid cash.
- What does the guaranteed column show at year three, and at a reduced dividend scale? This covers criticisms 3, 4 and 8 at once.
- Would this policy take dollars I might put into registered room, and who can advise me on that side? This covers criticism 6.
- Who lends on a policy loan, who receives the interest, and who sets the rate? This covers criticisms 2 and 7, and it separates someone who understands the mechanics from someone repeating a phrase.
- Who should not buy this, and how are you paid on it? This covers criticism 5. A straight answer names categories of people and states the compensation.
The reaction to the questions is as informative as the answers. A description that can carry its own strongest objections has told you what it is.
What if you think a policy was described to you inaccurately?
There is a route, and it works better when started early. Ask for a copy of your application and the illustration you were shown, and for the records the insurer relied on; a privacy access request can reach the rest of your file. Keep a copy of everything you send and receive.
Speak to a lawyer promptly, or in Quebec a lawyer or notary, because deadlines apply.
Who is making this case, and how is the firm paid?
Criticism of this approach comes from different places. Some comes from people who sell competing products and have an interest in the conclusion. Some comes from independent analysts with no stake at all. Some comes from households who bought a policy, were disappointed and are describing a real experience accurately. The source is worth knowing, and it settles nothing. Test each argument against the contract, the statute or the arithmetic.
Apply the same test here. Reading is free. The firm behind this site is paid by the insurer, by commission, if a policy is bought through it, as the author page states. The concessions above are published because they are true, and because a household that buys on an overstated case has no reason to stay when reality arrives, and leaving early is expensive in this product. Whole life insurance is insurance, not an investment, and a slow decision ages better than a fast one.
A thirty-minute discovery meeting
A first conversation establishes whether this fits. No illustration is prepared and nothing is arranged.
Often the answer is no, and you will hear it during the call rather than in a proposal afterwards.
This form reaches Canadian Wealth Creation Centre Inc. Any meeting, any advice and any insurance product is provided by Canadian Wealth Creation Centre Inc., through its representatives who are licensed in the client's province. IBC Financial is the company's educational website: it distributes no product and no financial service, and it gives no individualised advice.
Common questions
How many of the standard criticisms of this approach are correct?
Why does the comparison criticism matter so much?
Does interest on a policy loan come back to me?
Are policy loans taxable in Canada?
Is the guaranteed column in an illustration really guaranteed?
Does the insurance company keep my cash value when I die?
Is buying term life insurance and investing the difference better?
Should I fill my TFSA or RRSP before buying a policy?
Is the commission paid to the advisor a conflict of interest?
Is this approach a scam?
Do American criticisms of this approach apply in Canada?
Is the criticism of the vocabulary fair?
What happens if I stop paying premiums?
What facts should I have before I decide?
Who makes these criticisms, and does the source matter?
What can I do if I think a policy was sold to me on an inaccurate description?
Sources
- Income Tax Act, R.S.C. 1985, c. 1 (5th Supp.), subsection 148(1), paragraph 148(2)(d) and subsection 148(9) (definitions of disposition, paragraphs (b), (c), (f), (g) and (j), policy loan, and proceeds of the disposition), Justice Laws Canada, as read for this site's objections and risks hub, verified 2026-09-29
- Income Tax Act paragraph 60(s) (deduction for repayment of a policy loan, up to amounts previously included), Justice Laws Canada, as recorded on this site's policy loans page, verified 2026-09-16
- Income Tax Regulations, Regulation 306, Justice Laws Canada, verified 2026-08-21
- Canada Revenue Agency, Form T2210, Verification of Policy Loan Interest by the Insurer, verified 2026-09-28
- Assuris, whole life protection (up to $1,000,000 or 90% of the death benefit and up to $100,000 or 90% of the cash value, whichever is higher, calculated after deducting policy loans), verified 2026-09-29
- Autorité des marchés financiers, Comment utiliser une valeur de rachat sans mettre fin à son assurance (surrender value not payable on death; policy advance bears interest; borrowing from another institution with the policy as security), verified 2026-09-29
Last reviewed 2026-09-29. By Jose Salloum, Financial Security Advisor in Quebec. In Ontario, Life and Accident & Sickness Insurance Agent. In British Columbia, Life Insurance Agent.
Get Started
