Whole Life vs. Universal Life: Keep, Exchange, or Surrender a Policy

Whole Life vs. Universal Life: How Each Policy Works

Every permanent policy does the same three things: it charges for the insurance, it credits growth to the cash value, and it pays a death benefit. The designs differ in who carries which risk.

Whole life. The premium is fixed for life (or for a set number of years), and the policy guarantees a minimum cash value and death benefit at every age. A participating whole life policy also pays dividends, which are not guaranteed: they reflect the insurer’s actual investment results, mortality experience, and expenses against the conservative assumptions built into the guarantees. Dividends can be taken in cash, used to reduce premiums, or used to buy small paid-up additions of extra coverage that have their own cash value, which is how many older policies have grown well beyond their original face amount. The insurer carries most of the risk, and charges for it in the form of a premium far higher than term insurance for the same initial death benefit.

Universal life. The owner pays premiums into an account. Each month the insurer deducts a cost of insurance charge, equal to a rate set by the insured’s age multiplied by the net amount at risk (the death benefit minus the cash value), plus administrative charges, and credits interest to what remains. The current rates can be raised later, up to maximums the policy guarantees. Premiums are flexible: pay more and the cash value grows; pay less and the charges eat into it. The cost of insurance rises every year, slowly at first and steeply in the seventies and eighties, so a policy that was comfortably funded at 45 can be quietly running out at 75. Indexed universal life credits interest based on a stock index, within a cap and a floor; variable universal life invests the cash value in subaccounts. Both move investment risk to the owner, and the illustrations that sell them depend heavily on assumed returns.

Guaranteed versus illustrated. Every policy comes with an illustration, a year-by-year projection of premiums, cash value, and death benefit. The guaranteed columns show what the contract promises if the insurer’s experience is poor; the current columns show what happens if today’s dividend scale or crediting rate continues unchanged for decades. The current columns are not a forecast, and for most policies the gap between the two widens every year. An in-force illustration, which the insurer will run on request for a policy already owned, is the single most useful document in any policy review.

The insurer behind it. A guarantee is only as good as the company making it. State guaranty associations step in when a life insurer fails, but with limits that depend generally on where the owner lives: for Texas residents, up to $300,000 of death benefit and $100,000 of cash value per person for a failed insurer, and other states set their own. Above those limits, what a policyholder recovers depends on what the failed insurer’s remaining assets can pay, which is why the insurer’s financial strength belongs in every review alongside the numbers.

How Life Insurance Is Taxed: Cash Value, Loans, and the Death Benefit

Life insurance enjoys a set of tax advantages no other savings vehicle combines, and each comes with a condition that, if broken, turns the advantage into a bill.

The death benefit. Proceeds paid because of the insured’s death are generally free of income tax under Section 101. The exception that catches families is the transfer-for-value rule: a policy sold or otherwise transferred for value to someone outside a list of exempt buyers (the insured, a partner of the insured, a partnership with the insured as a partner, or a corporation in which the insured is a shareholder or officer) loses much of that exclusion. Gifts are generally exempt too. It matters most when policies change hands among business owners, a point our article on buy-sell agreements after Connelly covers.

The cash value. Growth inside the policy is not taxed as it accrues. The owner’s basis, which the tax code calls the investment in the contract, is the total premiums paid, less dividends taken in cash or used to reduce premiums and any earlier tax-free withdrawals. Dividends used to buy paid-up additions do not reduce it. For a policy that is not a modified endowment contract (below), withdrawals come out basis first, so they are tax-free until they exceed what was paid in (with an exception for some withdrawals in a policy’s first 15 years that come with a cut in benefits), and loans are not taxed at all while the policy stays in force. Loan interest paid by an individual is generally not deductible.

Surrender. Ending a policy for its cash value produces ordinary income equal to the amount received minus the investment in the contract. It is never a capital gain, even after decades. A policy whose cash value is below its basis produces no deduction for the shortfall when surrendered; that loss is simply gone.

Modified endowment contracts. A policy funded faster than a statutory limit (the seven-pay test, roughly the premium that would pay the policy up in seven level annual installments) becomes a modified endowment contract. It keeps the tax-free death benefit, but loans and withdrawals are then taxed income first, and taxable amounts taken before 59 1/2 generally carry an additional 10% tax. A policy can become one later, through a large additional premium or a material change in benefits, and a modified endowment contract exchanged for a new policy stays one.

Exchanges. Section 1035 lets an owner exchange a life insurance policy, without tax, for another life insurance policy, an annuity, or a qualified long-term care insurance contract, carrying the old policy’s basis into the new one. The exchange runs one direction only: an annuity cannot be exchanged into life insurance. Our article on how annuities are taxed covers what happens after a policy becomes an annuity. A policy loan that is paid off as part of an exchange is generally treated as cash received and taxed to the extent of the policy’s gain.

The estate. Life insurance is free of income tax but not of estate tax. A policy the insured owns, or controls through any incident of ownership, is included in the insured’s estate at its full death benefit under Section 2042. With the federal exemption at $15,000,000 per person in 2026, that matters to few families. For those it does reach, a policy owned from the start by an irrevocable life insurance trust stays out of the estate, while an existing policy transferred to one stays in the estate if the insured dies within three years.

Term or Permanent: Why We Lean Toward Term

Most life insurance needs end. Children grow up, mortgages are paid off, and a household that has accumulated enough to replace its own income no longer needs a death benefit to do it. Term insurance covers needs with an end date for a fraction of the cost of permanent coverage, because it is priced for the years the risk exists rather than for a lifetime.

That is why we lean so strongly toward term insurance and so rarely recommend buying a permanent policy. A permanent policy asks one contract to do two jobs, insuring a life and holding savings, and it does each less well than a tool built for that job alone. The savings carry the policy’s commissions and charges, which are hard to see. They can be reached only through a loan, a withdrawal, or a surrender taxed as ordinary income. And in most designs they are not paid on top of the death benefit: when the insured dies, the family receives the death benefit, and the cash value is part of what pays it. Keeping the two jobs apart, with term coverage for the years the need lasts and an investment account for the savings, leaves each one visible, flexible, and priced on its own.

Viewed on its own, the cash value of a well-funded whole life policy behaves like a conservative, insurer-backed, tax-deferred savings account, with returns that arrive only after the cost of the insurance and the policy’s expenses are paid. It does not fall in value when interest rates rise, as a bond fund can, which is why it is sometimes presented as a substitute for bonds. The deferral is real, but a new policy carries its highest costs in its early years, a policy surrendered in its first decade usually returns less than was paid in, any gain taken by surrender is ordinary income, and the savings rest on one insurer’s promise. For most households, the same money in an investment account does the job at lower cost and with more control.

The exceptions are narrow, and they share one feature, a need that does not end, or ends only at death:

• Business continuity. A buy-sell agreement among co-owners needs cash at a death, whenever it comes, for as long as the business is owned together. This is the case where permanent coverage is most often merited: term coverage can fund the agreement while the owners are younger, but as they age, a death benefit that will not expire before the agreement does is often the most practical funding. A policy the business itself owns on an owner or employee also needs the insured’s written notice and consent before it is issued, or much of the death benefit can become taxable to the business.
• Estate tax liquidity. A family whose estate will owe estate tax, often because of a business or real estate that cannot be sold quickly, can use a death benefit owned outside the estate to pay the tax without a forced sale. Our article on the $15 million estate tax exemption describes who still faces the tax, and our article on spousal lifetime access trusts shows insurance on a spouse protecting a household’s access to a trust.
• A lifelong dependent. A child or other family member who will need support for life, often through a special needs trust, is a need that does not end at retirement.
• Equalizing an inheritance. A family leaving a business or a ranch to one child can use a death benefit to give the others a comparable share when other assets cannot.

Even in those cases, we lean toward a policy chosen for the lowest cost of a lifelong death benefit rather than for its cash value.

How commissions work matters to that conversation. Permanent life insurance pays the selling agent a commission concentrated in the first year, often a large share of that year’s premium. That does not make any particular policy wrong, but it means a recommendation to buy a new policy, or to replace an old one, usually comes with compensation attached. A fee-only planner receives nothing from the insurer either way, a distinction our article on what a fee-only planner does explains. Our own incentive can run the other way: a surrendered policy’s cash value invested with us would add to an asset-based fee, which is one more reason the comparison belongs in numbers rather than impressions.

Reviewing a Policy You Already Own

The question most people bring is not whether to buy permanent insurance but what to do with a policy bought twenty or thirty years ago, often by a younger version of themselves facing a different problem.

The first step is the same as with any asset already owned: set aside what has been paid. Premiums, commissions, and early-year charges already spent are gone under every option, keeping and surrendering alike. What remains live is what each option does from here: the cash value available today, the premiums still ahead, the death benefit the family still needs, any surrender charge still in force, the tax basis, and the insurer’s strength.

That framing often runs against intuition in both directions. A whole life policy owned for twenty years has already paid for its most expensive years, so its forward return can be noticeably better than its history; dropping it now locks in the early costs without collecting the later growth. A universal life policy owned for the same twenty years may be in the opposite position, because its cost of insurance keeps rising, and the years ahead can be its most expensive.

The questions that decide it:

• Is the death benefit still needed, and how much of it? A policy is worth more to a family that would otherwise buy coverage, and less to one that would not.
• How is the insured’s health? The return on a death benefit depends on when it is paid, so poorer health makes keeping a policy more valuable, usually worth more to the family than a buyer would pay for it.
• What does the in-force illustration show from here, at both the current and guaranteed assumptions, and when, if ever, does the coverage end? For universal life, the year-by-year detail behind it (the cost of insurance charged, the expenses, and the interest credited) shows why.
• What could the cash value and future premiums do elsewhere, after tax?
• What is the investment in the contract, and what tax would each way out produce?
• Is there a loan, including one the owner may not know about? A whole life policy whose premiums stopped may have an automatic premium loan provision quietly borrowing to pay them.
• Who owns the policy? When an irrevocable trust owns it, the decision belongs to the trustee, who generally owes the trust’s beneficiaries a careful review before letting coverage go, subject to what the trust document says.
• How strong is the insurer, and how much of the cash value sits above the guaranty association limits?

A Worked Example: A Whole Life Policy at 58

A couple in their late fifties lives in San Antonio. When the husband was 35, he bought a participating whole life policy and has paid an $11,000 annual premium ever since: 23 years and $253,000 in all. Dividends have bought paid-up additions, so the death benefit is now about $1,050,000. The policy has no loans, the cash surrender value is $305,000, and his investment in the contract is the $253,000 he paid. The couple’s taxable income puts them in the 32% bracket.

The in-force illustration. Asked for a projection from today, the insurer shows that if he keeps paying $11,000 a year until 75, the cash value at 75 will be about $865,000 at the current dividend scale and $630,000 on the guaranteed values alone, with a death benefit at 75 of about $1,600,000 at the current scale.

What keeping the policy earns from here. Treat the decision as an investment made today: the $305,000 he could take now, plus $11,000 a year for 17 more years. At the current dividend scale, that grows to $865,000 by 75, a forward return of about 4.0% a year. On the guarantees, it grows to $630,000, about 1.8% a year. Both are tax-deferred, and if the policy is held until death, the growth is never taxed at all.

What surrendering earns. Surrendering today produces $52,000 of ordinary income ($305,000 minus $253,000), about $16,640 of federal tax at 32%, and $288,360 to invest. Suppose he invests it, along with the $11,000 a year he no longer pays in premiums, in taxable bonds yielding 4.5%. At their income, the interest also bears the 3.8% net investment income tax on top of the 32% bracket, so the bonds earn about 2.89% after tax. By 75 that account holds about $712,000. (The example taxes the policy’s gain at 32% alone; whether the 3.8% also reaches a surrender gain is less settled, and if it does, the policy’s lead shrinks to a few thousand dollars.)

The comparison. If he keeps the policy and surrenders it at 75 instead, the gain then is $425,000 ($865,000 minus his $440,000 of total premiums), the tax at the same flat 32% is about $136,000, and he keeps about $729,000. On the current dividend scale, the policy comes out about $17,000 ahead on cash; the bond would need to yield about 4.8% to match it. That margin is thinner than it looks, because the flat rate flatters it: a $425,000 gain taken in a single year at their current income would push much of it into the 35% and 37% brackets and erase most of the lead, while the same surrender in a lower-income year after retirement would cost less. On the guaranteed values, the policy leaves about $569,200 after tax, well behind the bonds. And the policy carries something the bond account does not: a death benefit of about $1,600,000 at 75 at the current scale.

The death benefit’s own return. Measure the death benefit the same way, against the $288,360 he would keep by surrendering today plus the $11,000 a year. If he dies at 75, the family receives $1,600,000 at the current scale, a return of about 8.6% a year, free of income tax. Even the death benefit already in force, about $1,050,000 (the base policy plus the paid-up additions bought so far, which the contract guarantees as long as premiums are paid and no loan is taken), works out to about 5.6% a year. The bond account at the same age holds about $712,000. Each additional year he lives adds premiums and pushes the payment further out, so the return falls with age, but the death benefit is always larger than the cash value and is paid free of income tax. That is why his health matters to the decision as much as the dividend scale does.

What decides it, then, is not the cash value. If the family wants the death benefit, for estate liquidity, a lifelong dependent, or simply to leave a larger, tax-free inheritance, keeping the policy buys it at no cost compared with bonds at the current scale, and at a real cost if only the guarantees hold. If they do not, the case for keeping rests on how much confidence they place in the dividend scale holding for 17 years, against a taxable alternative they control and could reach at any time. Either way, the history of the policy, the $253,000 already paid, plays no part in the answer.

The Options, and What Each Does to the Tax Bill

A policy owner who decides the current arrangement no longer fits has more choices than keep or surrender.

Keep paying. The policy continues as designed. Worth checking each year: the dividend scale, any loan, and whether the death benefit still matches the need.

Stop paying, keep coverage. Whole life policies carry nonforfeiture options set by state law. Reduced paid-up insurance uses the cash value to buy a smaller policy that needs no further premiums; in the example, the insurer’s illustration shows a paid-up death benefit of about $560,000. Extended term insurance uses the cash value to keep the full death benefit in force for a fixed number of years, after which coverage ends. In a policy with no loan, neither triggers tax, because nothing is received. With a loan outstanding, the conversion often settles the loan from the cash value, and that can be taxable to the extent it exceeds the basis. (In a policy still in its first seven years, a cut in benefits can also make it a modified endowment contract.) Dividends can also be redirected to pay premiums, if they are large enough.

Exchange it. A Section 1035 exchange moves the cash value into a new life policy, an annuity, or a qualified long-term care contract with no tax, and the $253,000 of basis in the example moves with it. Exchanging into a new life policy restarts its acquisition costs, requires underwriting at today’s age and health, and usually starts a new two-year contestability period, and an older policy may guarantee an interest rate no new policy will match, so a replacement has to clear a high bar. Exchanging into a long-term care or hybrid policy turns a death benefit nobody needs into care protection, as our article on funding long-term care describes. Exchanging into an annuity defers the gain, and for a policy whose cash value has fallen below its basis, it preserves that basis: the shortfall that a surrender would simply lose can instead shelter future growth inside the annuity from tax. It also changes the order in which money comes out. Withdrawals from a life policy come out basis first; withdrawals from a deferred annuity come out gain first, and taxable amounts taken before 59 1/2 generally carry an additional 10% tax. And an annuity gives up what made the policy valuable at death: its gain passes to heirs as ordinary income, with no step-up, where a death benefit would have arrived free of income tax. An exchange into an annuity fits best when the plan is to turn it into lifetime income, where each payment returns part of the basis tax-free, or when the policy is worth less than its basis and the annuity can grow under that basis.

Surrender it. The cash value arrives at once, and everything above the basis is ordinary income in that year, $52,000 in the example. A surrender in a lower-income year, after retirement but before required distributions begin, costs less than one in a peak earning year.

Sell it. A life settlement is a sale of the policy to an investor, who keeps paying premiums and collects the death benefit. It can pay more than the cash surrender value, usually for insureds in their late sixties or older, or younger ones in poor health, whose policies are worth more to a buyer than to the insurer. The gain is ordinary income up to the amount a surrender would have produced and capital gain above that. An insured a physician certifies as terminally ill, expected to die within 24 months, who sells to a licensed viatical settlement provider generally owes no income tax on the proceeds.

A sale also moves the death benefit to a stranger, sends the insured’s medical records to prospective buyers, and lets the buyer check on the insured’s health for life. In Texas, settlement providers and brokers must be licensed by the Department of Insurance, the broker’s compensation must be disclosed to the seller, and contact about the insured’s health is generally limited to once every three months. Offers vary widely, so it is worth more than one bid. And an offer above the cash value is itself a signal: if an investor can pay that, cover the transaction’s costs, and still expect a profit from a death benefit that is taxable to the investor above what it paid for the policy and in premiums since, the same policy is usually worth more still to a family that can afford the premiums and does not need the cash.

Give it away. A policy can be given to a child or to an irrevocable life insurance trust. The gift is valued at roughly the policy’s cash value plus unearned premium (technically its interpolated terminal reserve plus unearned premium, which the insurer reports on Form 712), and an existing policy given away stays in the insured’s estate if the insured dies within three years.

Universal Life: The Policy That Can Quietly Run Out

Universal life policies bought in the 1980s, 1990s, and early 2000s were often illustrated at interest rates far higher than they have actually credited since, with premiums set at the minimum the illustration said would carry the policy for life. When credited rates fell, those premiums stopped being enough, and the gap has been made up by drawing down the cash value, which in turn raises the amount at risk and the cost of insurance charged against it. Some insurers have also raised the current cost of insurance rates on older policies, within the guaranteed maximums. Many owners find out only when the insurer sends a notice that the policy will lapse without a much larger payment, often in their late seventies, when the cost of keeping it is highest and replacing it is hardest.

Three steps head that off. First, request an in-force illustration at the current crediting rate and at the guaranteed rate, showing the age at which coverage ends if premiums continue as they are, along with the year-by-year ledger of charges and interest credited behind it. Second, ask what level premium would carry the policy to age 95 or beyond at current assumptions; that number, set beside the death benefit, is the real price of keeping the coverage. Third, decide while there is still cash value to work with: a universal life policy can usually be reduced to a smaller death benefit that the existing cash value and premiums can sustain, exchanged, or, for an insured with health issues, sold. For that same insured, paying more to keep the full death benefit can also be the strongest choice, because a death benefit likely to be paid sooner earns a high return on the added premium. A policy discovered at 79 with nothing left offers far fewer choices.

Indexed and variable universal life add one more question: what return the illustration assumes. A policy illustrated at a rate the index or subaccounts have not delivered faces the same slow drain, and the cap rates on indexed policies can change after purchase.

Policy Loans and the Lapse Trap

A policy loan is not taxed when taken, which makes it a popular way to draw on cash value. It is still a loan, made by the insurer with the cash value as collateral, and however it is described, the interest is paid to the insurer, not to the owner. Interest accrues, usually at a rate set by the policy, and unpaid interest is added to the balance. The cash value keeps earning while it secures the loan, though in some whole life policies the borrowed portion earns a different dividend rate than the rest (called direct recognition), which changes the loan’s true cost. As long as the policy stays in force, nothing is taxed, and at death the loan is simply subtracted from the death benefit.

Not every loan is deliberate. Many whole life policies include an automatic premium loan provision: if a premium goes unpaid, the insurer pays it with a loan against the cash value. An owner who stopped paying years ago, believing the policy had become self-sustaining, may own a policy that has been borrowing against itself ever since.

The trap is a lapse. If the loan balance grows until it equals the cash value, the insurer sends a notice, and once the grace period runs out the policy terminates, and the outstanding loan is treated as an amount received on surrender. The owner owes ordinary income tax on the difference between that amount and the investment in the contract, in a year when no cash arrives at all. The Tax Court has upheld that result repeatedly, including in a 2025 decision.

The arithmetic is unforgiving. Take a policy with a $400,000 loan balance that has grown to equal its cash value, and an investment in the contract of $200,000. If it lapses, the owner has $200,000 of ordinary income and, at a 32% rate, about $64,000 of tax to pay from other money, with the coverage gone.

An overloaned policy needs a decision before the insurer makes one. The choices include:

• Paying down the loan from outside money. Cash or bonds earning less than the policy’s loan rate are often the cheapest source.
• Paying the interest each year, from outside money or by redirecting dividends to it, so the balance stops growing.
• Surrendering some paid-up additions to pay down the loan. In a policy that is not a modified endowment contract, that comes out basis first, though it lowers the cash value by as much as the loan, so it slows the drain mainly when the loan rate is higher than what the additions earn.
• Reducing a universal life death benefit, so the cost of insurance charged against the cash value falls. (Converting an overloaned whole life policy to reduced paid-up insurance often settles the loan from the cash value, which can produce much of the same tax a lapse would.)
• Keeping the policy in force until death, when the loan is repaid from the death benefit and no income tax is due.

Exchanging an overloaned policy into an annuity does not escape the gain, because a loan paid off as part of the exchange is treated as cash received. Some life insurers accept an exchange that carries the existing loan into a new policy, which keeps the deferral, though the new policy deserves the same scrutiny as any replacement. Repaying the loan from outside money before any exchange avoids the question.

How We Think About It

We lean strongly toward term insurance, and we rarely recommend buying a permanent policy. Insurance should do the insurance job and investment accounts the investing job; a policy asked to do both does each at a higher cost, with charges that are hard to see and savings that can be reached only through a loan, a withdrawal, or a taxable surrender. The exceptions are needs that truly last for life, most often funding a buy-sell agreement among co-owners as they age, and less often estate tax liquidity for an estate that will owe it, support for a lifelong dependent, or an inheritance that cannot be equalized another way. Even then, we lean toward a policy chosen for the lowest cost of a lifelong death benefit rather than for its cash value.

A policy already owned is a different question from buying one. We lean toward judging it only by what it does from here: its forward return at the current and guaranteed scales, the death benefit the family still needs and the return it earns given the insured’s health, the tax on each way out, and the insurer behind it. Premiums and charges already paid are gone under every option, and basis is a live input, especially for a policy worth less than was paid in, where an exchange can preserve a loss a surrender would waste. For a whole life policy owned for decades, the most expensive years are usually behind it and the forward math can be better than its history suggests; for a universal life policy whose charges rise with age, it can be worse. When a household keeps a policy, we lean toward counting its cash value with the household’s conservative assets when the overall mix of investments is set, keeping in mind that it rests on one insurer’s promise and is slower to reach than cash.

We lean toward treating a policy loan as debt against the death benefit, reviewed against the cash value every year (including any automatic premium loan the owner did not choose), and toward resolving an overloaned policy, by repaying, reducing, or planning around the tax, before a lapse decides it. A policy review belongs in the same multi-year plan as the rest of a household’s income and taxes, because the year a policy is surrendered, exchanged, or lapses is a tax event like any other.

Common Questions

What is the difference between whole life and universal life insurance?

Whole life has a fixed premium, guaranteed minimum cash values and death benefit, and, in participating policies, dividends that are not guaranteed. Universal life has flexible premiums: the owner pays into an account, the insurer deducts expenses and a cost of insurance charge that rises with age, and credits interest to what remains. Whole life puts more of the risk on the insurer; universal life puts more on the owner, and an underfunded universal life policy can lapse late in life.

Should I keep or surrender my whole life policy?

Judge it by what it does from here, not by what has been paid in. An in-force illustration shows the forward growth of the cash value at the insurer’s current and guaranteed assumptions; compare that with what the cash value and future premiums could earn elsewhere after tax, and weigh whether the death benefit is still needed and what it is worth given the insured’s health. A surrender makes the gain above the premiums paid ordinary income, and alternatives include a reduced paid-up policy, a tax-free exchange, or a sale.

Is the cash value of life insurance taxable?

Growth inside the policy is not taxed while it stays in force, and withdrawals from a policy that is not a modified endowment contract come out of the premiums paid first, tax-free. Loans are not taxed while the policy stays in force. A surrender, or a lapse with a loan outstanding, makes everything above the premiums paid ordinary income, and the death benefit is generally free of income tax.

Is whole life insurance a good investment?

For most households, we do not think so. A whole life policy combines insurance with a conservative savings account, and the savings carry the policy’s commissions and charges, can be reached only through a loan, a withdrawal, or a surrender taxed as ordinary income, and in most designs are part of what pays the death benefit rather than an addition to it. We lean toward term insurance for the death benefit and an investment account for the savings, and toward permanent coverage only for the rare need that lasts for life. A policy already owned for many years is a different question, judged by what it will do from here rather than by what has been paid in.