Search for the problems with indexed universal life insurance and you’ll land in a shouting match: caps that get slashed, illustrations that lie, policies that implode, and the big one — eight out of ten IUL policies supposedly fail. Most of these warnings describe something that can happen. Almost none of them tell you how often it actually does. That gap is the whole story.
The Short Version
The five biggest IUL worries, and where the evidence actually lands
- Cap-cutting and index complexity: the fear that carriers gut your cap is mostly a fear — but the gap between what an engineered index promises in a backtest and what it delivers live is real, measured, and the strongest concern of the five.
- Illustration gaming: regulators have rewritten illustration rules three times in a decade and product design keeps routing around them. It’s a real problem — but a point-of-sale problem, not proof that policies are failing.
- Underfunding and rising cost of insurance: a genuine mechanism with zero incidence data. What actually matters is the ratio of premium to death benefit, not the crediting rate.
- The loan-lapse tax trap: real, litigated, and completely unquantified. It’s almost entirely a matter of whether anyone monitors the policy.
- “8 out of 10 policies fail”: unsourced and arithmetically implausible. The most-repeated claim against IUL is also the weakest.
Here’s a fact worth sitting with before you read a single scary headline. In 2025, IUL new premium hit a record $4.5 billion, up 17% over the prior year, with roughly a 25% share of the entire U.S. individual life insurance market. A product that was quietly destroying the people who own it, at the rate its critics imply, would be shrinking. It’s doing the opposite.
That isn’t proof of anything on its own — sales measure how hard a product is being distributed, not how well it treats its owners. But it’s the right place to start, because the case against IUL has a peculiar shape. It’s unusually rich in mechanism and unusually poor in incidence. Critics can describe, in accurate detail, exactly how an IUL policy fails. What almost nobody produces is evidence of how often it has. If you’re still getting your bearings on the product itself, our plain-language explainer on what an IUL policy is is the place to start; this post is about the criticisms.
Five Worries, One Pattern
We worked through the five biggest complaints about IUL on the podcast, and for each one we asked two separate questions that usually get blurred together. First: is the mechanism real — can this happen? Second: is there incidence data — do we know how often it happens? The answers don’t line up the way the doom narrative assumes.
| The worry | Mechanism real? | Incidence data? | Where it lands |
|---|---|---|---|
| Index complexity & cap control | Yes | Partial, and damning | Strongest of the five — the backtest-vs-live gap is measured |
| Illustration gaming | Yes | Yes (regulatory record) | Strong, but a sales-practice problem, not a policy failure |
| Underfunding & rising cost of insurance | Yes | None | Real risk, zero measurement |
| Phantom income from a loan-driven lapse | Yes (settled tax law) | None | Litigated, unquantified |
| “8 out of 10 policies fail” | Partly | Very thin | Weakest of the five — the headline claim is unsourced |
The through-line: almost every worry describes something that can go wrong, and almost none of them tells you how often it does.
Worry #1: The Caps and the Complexity
This worry has two halves that get conflated, and pulling them apart is where most of the clarity lives. The first half is carrier discretion: caps, participation rates, and spreads aren’t guaranteed, so the insurer can lower your crediting potential after you’ve already committed years of premium. The second half is index construction: a growing share of IUL crediting is tied not to the plain S&P 500 but to proprietary volatility-controlled indices — engineered indexes, built by investment banks, marketed with backtested histories that predate the index’s own existence.
The cap-cutting fear is mostly a fear
The story goes that carriers quietly grind your cap down toward the contractual minimum just to keep more money. It’s technically possible. But there are two problems with it as a description of reality. First, the money in the index account isn’t the insurer’s to keep — and insurers are legally restricted from speculating with options in the general account; options exist to hedge, not to harvest gains. Second, caps track interest rates and volatility with striking consistency. When rates fell and volatility spiked, caps came down. When rates recovered, caps went back up — something we’ve actually watched happen. Carriers don’t control interest rates or volatility any more than you do.
Are there carriers that have been slower and stingier than they needed to be about restoring caps? Yes. There are winners and losers in every product line. But “a carrier managed its caps less generously than a competitor” is a very different claim from “carriers arbitrarily confiscate your upside,” and only the first one is supported by anything we can find.
The index-construction problem is the real one
This is the single best-documented failure in the entire IUL critique, and it deserves to be taken seriously. When an index is built by fitting rules to a historical dataset, those rules will describe that history beautifully. That’s not a prediction — it’s a curve. The question is what happens once real money is trading against it.
An analysis published in InsuranceNewsNet in April 2024 looked at twelve commonly used volatility-controlled indices and compared their live performance to the S&P 500 since each one’s inception. Every single one captured less than 40% of the S&P 500’s return. Eight of the twelve captured less than 10%.
Take one index as the clean worked case. The Credit Suisse Balanced Trend 5% Index launched in November 2017. Its backcast — run from 2002 up to the launch date — produced 6% annualized against 7.6% for the S&P 500, a 78% capture ratio and the number a client would have been shown. Once it was live and trading with real money, it captured 9.1% of the S&P 500’s growth.
Backtest vs. reality
Credit Suisse Balanced Trend 5% Index — share of the S&P 500’s return captured
Source: Nick Pratt, InsuranceNewsNet, April 2024. The backtest promised 78% of the S&P 500; live performance delivered 9%.
The gap gets more pointed when you look at the specific years these indices were engineered to handle. The backcast said Balanced Trend would have gained 0.8% in 2008 while the S&P 500 lost 38.5%. But live, in 2022, it lost 11.1% while the S&P 500 lost 19.4% — it participated in exactly the downside it was designed to avoid. Twelve for twelve underperforming isn’t bad luck; it’s what you’d expect from a curve fit to history rather than a strategy that works going forward.
The evidence verdict — strongest of the five
On index construction, the critics are right and they have the receipts. On carrier discretion, they have a plausible fear and almost no evidence of the confiscation scenario they describe. The practical takeaway is narrow and actionable: if you’re going to worry about one thing here, worry about which index your policy is tied to, not whether the carrier will someday cut your cap. If the index didn’t exist independently before the insurance product that references it, treat its history as marketing. A plain S&P 500 cap-and-floor structure doesn’t have this problem — there’s a century of live history and nobody designed the index to sell insurance.
For how these newer indexes actually work and where they fit in a modern policy, see our deeper piece on volatility-controlled indices.
Worry #2: The Illustration Game
An illustration is the voluminous document that projects how a policy’s cash value and death benefit unfold over time, based on an assumed crediting rate. In the early days of IUL, some of those assumptions were genuinely crazy — a static rate near 8%, a loan taken at a far lower rate, and a giant spread between the two that made magical things happen on a spreadsheet. As we’ve said for years, anything works on a spreadsheet if you put the right numbers in. That doesn’t mean it works in real life.
The criticism worth engaging isn’t that agents got carried away. It’s structural: illustration regulation is always one cycle behind product design, so the number on the page keeps drifting away from what the product can reasonably deliver — and it does so legally. The striking part is that this critique comes from the actuarial profession describing its own industry. A Society of Actuaries product-development article put it plainly: “Innovation tends to outpace regulation, and once again, developments in product designs not captured by the regulation led to continued inconsistency amongst IUL illustrations.”
Show me a rule and I’ll show you the workaround
The pattern is documented across three regulatory cycles. In 2015, Actuarial Guideline 49 capped the illustrated rate off a benchmark index account. Carriers responded with multipliers and bonuses layered on top of the capped rate, restoring the values the guideline was meant to constrain. In 2020, AG 49-A closed the multiplier gap. Carriers pivoted to proprietary volatility-controlled indices, which are cheaper to hedge — freeing up an option budget that gets recycled into fixed-account bonuses the guideline didn’t reach. In 2023, AG 49-B arrived to solve the volatility-controlled-index problem. It didn’t fully close it, and as of 2026 regulators are on a fourth draft.
None of that is carriers breaking the rules. It’s product design routing around each new rule within a few years, every time — the same thing that happens in every regulated industry. Which is exactly why you should read an IUL illustration as the start of a conversation, not a promise.
The evidence verdict — strong, but a sales-practice problem
This concern is well evidenced and we shouldn’t soften it. Three complete cycles of rule-then-workaround are documented in actuarial and regulatory sources, and regulators are on their fourth attempt more than a decade in. But be precise about what it proves. It proves that illustrations have been unreliable and that a regulatory ceiling too often got sold as a floor. It does not prove that in-force policies are failing. The harm here is expectation-setting at the point of sale — a real harm, and a different one from a policy that collapses.
This is why we stress-test every policy at reduced assumptions. An illustration at 7.5% looks spectacular; one run at 5.5% tells you what happens when markets don’t cooperate. The worst outcome isn’t a single bad year of crediting — it’s a policy sold on a number it was never going to hit.
Worry #3: Underfunding and the Rising Cost of Insurance
This is the most mechanically sound worry in the set and the one with the least data behind it — which makes it the most interesting. The mechanism isn’t in dispute by anyone. Cost of insurance in a universal life chassis is charged monthly against the amount at risk, and the per-thousand rate climbs with age. Early on it’s trivial. Late, it is not.
Cost of insurance climbs steeply with age
Annual mortality cost per $1,000 at risk — up roughly 40x from 45 to 90
Source: SSA Period Life Table, 2023 (raw male mortality). Actual policy charges are loaded above raw mortality; the slope is the point, not the level.
The critique layers a second claim on top of that curve, and this one is simply true: agents are compensated on the death benefit put in force, not on how well-funded the policy is. That structurally rewards maximizing face amount for a given premium — the exact opposite of what makes a cash-value policy durable. The number that matters isn’t the premium and it isn’t the crediting rate. It’s the ratio between the premium and the face amount.
Hold the premium constant, hold the crediting rate constant, change only the death benefit, and you get completely different policies. A million-dollar death benefit funded with a $2,500 annual premium isn’t a problem today — it’s a problem later, when the rising cost of insurance on all that death benefit outruns the cash value. Give that same premium a right-sized death benefit and it can carry for life. Structure dominates performance by roughly an order of magnitude, which means most of the industry’s decade-long argument about whether the index credits 5.5% or 4.5% is arguing about the second-order term. We put numbers to exactly this in when IUL underperforms whole life.
The evidence verdict — real risk, zero measurement
The mechanism is real and the incentive misalignment is real. But there’s no incidence data of any kind — not thin evidence, none — because “cause of lapse” simply isn’t a field the industry collects. No study anywhere attributes a policy lapse to cost-of-insurance escalation. Our honest position: this is the concern we’d most want data on, we can’t get it, and the absence of data isn’t reassurance. It’s a measurement failure. Anyone who tells you they know how many IUL policies failed from underfunding is guessing.
One adjacent datapoint runs the other way. Milliman’s industry review found that original pricing for universal life with secondary guarantees often assumed higher lapse rates than actually occurred — policyholders held on more tenaciously than the actuaries expected. That’s a different slice of the market than accumulation-focused IUL, but it’s a useful corrective. And the market keeps voting with its feet: Milliman notes that universal life — particularly IUL and variable UL — grew to a combined 42% of the business in 2024, up from 30% in 2019.
Worry #4: The Phantom Income Tax Trap
Of the five, this is the one with actual adjudicated cases, and the one that most deserves to be explained slowly, because the outcome genuinely surprises people. The IUL loan strategy is sold as tax-free income: borrow against cash value, never repay, let the death benefit settle the loan. That works — right up until the policy lapses while the loan is still outstanding.
Here’s the mechanism. When a policy terminates, the outstanding loan balance is treated as an amount received by the policyholder. The taxable gain is that amount less your basis — roughly the premiums you paid. But you already spent the loan proceeds, possibly a decade ago. So the tax bill arrives with no cash attached, in the same year the death benefit disappears, and there’s no policy left to borrow from to pay it. It’s not that the tax is unfair. It’s that the timing is catastrophic.
| Loan-driven lapse — worked example | Amount |
|---|---|
| Cumulative premiums paid (basis) | $250,000 |
| Cash value at the point of lapse | $412,000 |
| Outstanding loan balance | $408,000 |
| Net surrender value to the policyholder | $4,000 |
| Amount treated as received | $408,000 |
| Less basis | ($250,000) |
| Taxable gain reported on Form 1099-R | $158,000 |
| Federal tax at a 24% marginal rate | $37,920 |
| Cash actually received in the year of the tax bill | $4,000 |
Hypothetical example for illustrative purposes only. Individual results vary based on specific products, funding, timing, and personal circumstances. Read the last two lines together: a $37,920 liability against $4,000 of cash, taxed as ordinary income.
This isn’t theoretical. The Tax Court has ruled on it consistently and against the policyholder — Doggart, Sawyer, Mallory, and Fugler among them. In one of those cases the taxpayers attached a note to their return saying no one knew how to compute the liability from the 1099-R and the IRS couldn’t help when called. No leniency was granted. That’s the human texture of this failure mode, and it lands far harder than a citation to a tax code section. If you want the mechanics of the form itself, we broke it down in how to calculate the taxable amount on a 1099-R for life insurance.
The evidence verdict — real, litigated, and completely unquantified
This is the only one of the five with adjudicated outcomes, so the mechanism isn’t merely theoretical — it has claimed identifiable people. But there’s zero frequency data. No dataset tracks how many policies lapse with loans in excess of basis. The honest reading of the litigation record is that it tells us the trap exists, not how often it springs. And here’s the practical part: this risk is almost entirely a function of whether anyone is monitoring the policy in the loan years. It’s a service failure dressed up as a product failure — and it’s the single strongest argument for an annual in-force review that we have.
The risk is real and it’s manageable, but it has to be understood and it has to be watched. If you’re using — or considering — policy loans against life insurance, the person you hand that monitoring responsibility to matters a great deal. Ask hard questions before you move forward, not after.
Worry #5: The “8 Out of 10 Policies Fail” Claim
You’ve seen the number: eight out of ten IUL policies fail within twenty years, usually attributed to LIMRA. It circulates in social media threads, in term-insurance sales material, and in a lot of otherwise reasonable financial commentary. It’s the most-repeated statistic in the entire anti-IUL case. It’s also the one where the critics have overreached the furthest.
We went looking for it specifically, and it isn’t in any publicly available LIMRA study. Before you even get to the sourcing, it doesn’t survive a basic sanity check. Observed annual lapse rates across the universal life family run in the single digits. Compound single-digit annual attrition over twenty years and you land somewhere around 40% to 60% of policies persisting — not 20%. To get to an 80% failure rate you’d need sustained annual lapse rates near the very top of the observed range, every year, for two decades, with no improvement. The claim isn’t just unsourced; it’s arithmetically strained.
What the actual study says
The relevant primary source is the SOA Research Institute and LIMRA universal life lapse study covering 2015–2021, published in late 2023. It’s a serious piece of work: 24 contributing companies representing roughly 80% of industry sales, 33.5 million policy exposures, $8.5 trillion of face amount, and 1.3 million lapse terminations. Two of its findings cut directly against the doom narrative. First, in its own words, lapse rates “by count and amount have generally decreased relative to the prior study covering 2009–2013.” Rates are declining, not rising. Second, accumulation-focused indexed UL — the product people actually mean when they criticize IUL — is a very small slice of the study’s exposure, drawn from only a handful of contributing companies. The dataset that supposedly proves an 80% failure rate can barely see accumulation IUL at all.
There’s a deeper problem for anyone claiming IUL persistency has deteriorated: the prior 2009–2013 study couldn’t separate indexed UL from traditional UL at all. There’s no historical IUL baseline to compare against. Anyone asserting that IUL persistency has gotten worse is comparing a number to nothing.
The evidence verdict — weakest of the five
The headline claim is unsourced and arithmetically implausible against observed lapse rates. The best available study can’t resolve the question in either direction for accumulation IUL, because the relevant cell is a sliver of its data. Meanwhile the study’s broad findings run against the doom narrative: lapse rates generally decreased versus the prior study. This is the concern where we’re most comfortable saying the critics have overreached.
Where This Leaves Us
Line the five up and a pattern appears. The critique of IUL is strongest exactly where it addresses how the product is described at the point of sale, and weakest exactly where it addresses how the product has performed for the people who own it. That’s a coherent finding, and it’s not a defense of IUL.
The first two worries are the credible half. Backtested indices really did fail to deliver, twelve for twelve. Illustration regulation really has been outrun three separate times, by the actuarial profession’s own account. Both of those are about the number a client was shown before they signed. The last three are the half where the critique runs out of evidence: the underfunding mechanism is real and unmeasured, the tax trap is real and unmeasured, and the persistency claim rests on a number no published study supports.
The practical conclusion we can defend: the risks that have actually harmed identifiable IUL owners are overwhelmingly structural and behavioral — face amount set too high for the premium, allocation left unmonitored, loans left unreviewed, premiums quietly stopped — rather than product-mechanical.
A word on the headlines
You’ve probably seen the Kyle Busch–Pacific Life lawsuit and likely drawn the wrong conclusion from it. The NASCAR driver and his wife sued Pacific Life in October 2025 over five IUL policies, claiming losses exceeding $8.5 million, and the case settled confidentially in early 2026. But the policies were allocated essentially entirely to the fixed account, crediting around 2.25% — the cash value never had meaningful index exposure. Pacific Life’s defense argued the policyholders failed to timely pay planned premiums and failed to monitor their allocations. Strip away the celebrity and the enormous face amount, and it’s a funding, allocation, and disclosure dispute. It isn’t evidence that index crediting doesn’t work. We walked through the details in our breakdown of the lawsuit — and every piece of IUL litigation we’ve examined points the same direction: agents doing things they never should have, not carriers manufacturing a defective product.
The line to land on
Every one of these five worries describes something that can go wrong. Four of the five come with no evidence about how often it does. That’s not the same as saying nothing goes wrong — it means the industry doesn’t measure what happens to its own policyholders, and the critics have filled that vacuum with numbers they can’t source. Both are problems. Neither is an argument for or against owning the product. The right question was never “is IUL good or bad.” It’s whether this policy is designed, funded, and monitored the way it needs to be.
If you want the balanced, point-by-point version of the broader debate, we took apart one of the most-cited IUL critiques in a critical review of a critical review of indexed universal life insurance.
Common Questions
Are indexed universal life insurance policies a bad idea?
Not inherently — but a poorly designed one can be. The evidence shows the real harm to IUL owners is overwhelmingly structural and behavioral: a death benefit set too high for the premium, an index allocation left unmonitored, or policy loans left unreviewed. Those are design and service failures, not flaws baked into the product. A well-designed, properly funded, actively monitored IUL policy is a very different thing from the horror stories, and the difference is almost entirely in how it’s built and managed.
Can my insurance company lower the cap rate on my IUL policy?
Yes, it’s contractually possible — caps, participation rates, and spreads aren’t guaranteed. But there’s no documented case of a major carrier grinding an in-force block down to its contractual minimum. Caps track interest rates and volatility closely: when rates fell and volatility spiked, caps came down; when rates recovered, caps went back up. The only crediting number that’s guaranteed is the floor, so the more useful thing to scrutinize is which index your policy is tied to, not whether the carrier will someday cut your cap.
Is it true that 8 out of 10 IUL policies fail?
No — that figure isn’t in any publicly available LIMRA study, and it doesn’t survive basic arithmetic. Observed annual lapse rates across the universal life family run in the single digits, which compounds to roughly 40% to 60% of policies persisting over twenty years, not 20%. The most serious industry lapse study, from the SOA Research Institute and LIMRA, actually found that lapse rates have generally decreased relative to the prior 2009–2013 study. It’s the most-repeated statistic in the anti-IUL case and the least supported.
What is the “phantom income” tax trap on an IUL policy loan?
It happens when a policy that’s been borrowed against lapses while the loan is still outstanding. At that point the outstanding loan balance is treated as an amount received, and the gain above your basis is taxable as ordinary income — even though you spent the loan proceeds years earlier and there’s no cash left in the policy. The tax bill arrives with no cash attached, in the same year the death benefit disappears. The Tax Court has consistently ruled this way, and the risk is almost entirely a function of whether anyone is monitoring the policy in the loan years.
Why do IUL illustrations look too good to be true?
Because illustration regulation has consistently run one cycle behind product design. Each time regulators cap what an illustration may show — AG 49 in 2015, AG 49-A in 2020, AG 49-B in 2023 — product design finds a legal way around it within a few years, and a fourth revision is in progress. The Society of Actuaries has said as much about its own industry. The practical takeaway: read an illustration as the start of a conversation, not a promise, and always ask to see it run at a reduced crediting assumption.
Did the Kyle Busch lawsuit prove that IUL doesn’t work?
No, and it’s widely misread. The Busch policies were allocated essentially entirely to the fixed account, crediting around 2.25% — the cash value never had meaningful index exposure, so it isn’t an index-crediting story at all. Pacific Life’s defense argued the policyholders failed to timely pay planned premiums and failed to monitor their allocations. Strip away the celebrity and the enormous face amount, and it’s a funding, allocation, and disclosure dispute — not evidence that index crediting fails as designed.
What’s the single biggest risk of owning an IUL policy?
Underfunding relative to the death benefit. The cost of insurance in a universal life chassis rises steeply with age — roughly forty times higher per thousand dollars of coverage at age 90 than at 45 — so a policy with too much death benefit for the premium can be fine for years and then erode as those costs climb. Structure dominates performance by roughly an order of magnitude, which means getting the premium-to-death-benefit ratio right matters far more than whether the index credits 5.5% or 4.5%. This is also why an annual in-force review is worth doing.
Wondering if your IUL is designed correctly?
Most of the real risk in an IUL policy lives in how it’s built, funded, and monitored — not in the product itself. We review existing policies and design new ones. A 30-minute call is enough to see where you stand. No sales pitch.
Schedule a 30-minute call
or Prefer to write? Send us a message
This article is general education, not a recommendation for any specific product. Product suitability depends on individual circumstances including age, health, income needs, time horizon, and existing assets. We specialize in cash value life insurance and fixed annuities and do not sell or advise on securities-regulated products.
