Everyone knows insurance salespeople earn commission. That fact on its own explains nothing, and treating it as an accusation is how people end up refusing products they genuinely need.
The thing worth understanding is the shape of the payment, not its existence. Across several product families, most of what the seller earns arrives in the first year, and almost nothing arrives later.
That single design decision explains most of what goes wrong, because it severs the seller's payoff from whether the product still works for you in year eleven. Once the first-year money is banked and renewals are small, the highest-value action available with an existing client is not to keep serving them. It is to sell them something new.
Everything below follows from that.
State the exception first, because a pattern applied everywhere stops being informative.
Home and auto insurance does not work this way. Commissions are level rather than front-loaded, and the economics are visibly those of insurance: across the whole US property and casualty industry, direct claims came to 65.5% of premium in 2023 and 61.8% in 2024.1 Over the decade to 2023 the industry's average return on net worth was 6.0%, against 14.9% across the Fortune all-industry group.2
That is not a distribution scheme. Whatever is wrong with home and auto insurance — and plenty is — it is not that the premium mostly stays with the seller.
Before the patterns, the numbers — because this is the part of the subject that is hardest to find out, and the reason it is hard is itself worth knowing.
Commission rates on life insurance are not published anywhere. There is no register. They reach the public record almost entirely by accident of a different law: variable life insurance is a registered security, so insurers must file their selling agreements as registration exhibits. Term insurance, traditional whole life and indexed universal life carry no such requirement, and their rates stay private contract terms.
One insurer's career agent contract, filed only because it had to be, is headed "strictly confidential and none of its terms are to be disclosed without the company's consent."17 Another filed its term schedule with every percentage blacked out as a trade secret.17
From the schedules that did reach the public record:
Universal life. One agent-level schedule pays 90% of the target premium in the first year, 3% on anything above target, 2% in years two to ten, 1% from year eleven, and a 0.10% annual trail on assets. Then a separate wholesaling allowance adds 24% of target premium in year one — taking first-year distribution cost to 114% of the target premium.17
More than the whole first year's premium, paid out in the first year.
Across the filed schedules the first-year rate runs from 8% to 99% of target premium. The low end is institutional and low-load design; one insurer disclosed a ceiling of 99%.17
Whole life. 55% of first-year premium, 7% in years two to ten, and 3% on the paid-up additions rider.17
Renewals stop. Every schedule read shows the same shape — a step down after year one, then a cliff. The cut-off after year ten is close to universal.17
Only one state caps any of this. New York limits an agent or broker to 55% of qualifying first-year premium, 63% for a general agent, with renewals capped at 22%, 20% and 18% in years two, three and four — and, importantly, caps commission plus expense allowance together at 91%.17 That last provision exists because the allowance is how the cap gets exceeded everywhere else.
The most striking mechanism in the subject, and it is a named actuarial technique rather than a theory.
29% of permanent life policyholders lapse within three years, and 57% within ten. Nearly 88% of universal life policies never terminate in a death claim. Between 1990 and 2010 about $24 trillion of coverage was dropped — roughly 78% of all coverage issued.3
Insurers' own actuaries have shown what this does to profitability. One secondary-guarantee universal life policy was projected at minus 12.8% profit assuming nobody lapsed, and plus 13.6% at a typical 4% lapse rate. A 30-year term policy was worth plus $103,000 in present value under normal lapse patterns and minus $942,000 with none.3
The product needs you to stop doing the thing you bought it for. That is what "lapse supported" means.
The regulatory record is the evidence here, and it is unusually clear.
Indexed universal life policies are sold on an illustration — a projection of what the policy might do. Regulators have now made three successive attempts to constrain how those projections are produced: a guideline in 2015, a replacement for policies sold from 2020, and a further revision for new business from May 2023, with additional consumer-disclosure changes in 2026 and the standard-setting body still working on it as of May 2026.4
A decade of patches on one document is not evidence of an industry failing to comply. It is evidence of product design routing around each constraint as it arrives.
The general form of the pattern is worth carrying beyond insurance:
Products you can compare on a single number pay out most and cost least. Products you cannot compare pay out least and pay the seller most.
Level term against indexed universal life. An immediate annuity against a bonus indexed annuity. A plain homeowners quote against a home warranty contract. The relationship is consistent, and it is not perishable the way a rate table is.
The clean version of this is in the point-of-sale data. Extended warranty attachment ran 28.7% in store against roughly 4% online for the same goods.6
Same product, same risk, same price. What changed was whether the buyer could walk away and compare — and it changed the take-up sevenfold.
The pattern's sharpest case is the one where the number is missing. GAP is sold in the five minutes after a car price is agreed, and no American regulator publishes what it pays back in any channel. What is measured is the price: the Consumer Financial Protection Bureau put the average cost of GAP financed into a US auto loan at $952.5 A product sold where nobody can shop is also, here, a product nobody measures.
Consumer Reports puts the retail mechanics plainly: retailers push service plans "because they're cash cows for them. Stores keep 50 percent or more of what they charge for these contracts."7
If compensation resets on every new contract, then moving an existing client from one product to another regenerates a first-year commission. The client pays twice — a surrender charge going out, a fresh schedule coming in.
This is current, not historical. Four enforcement actions on the supervision of annuity exchanges landed within six months:8
an average incremental cost of $8,718.86 each
22 exchanges, 14 customers
The Ameriprise finding was a failure to supervise rather than a finding about any individual recommendation. That distinction matters and should not be flattened.
Regulators have also acted repeatedly on credit card add-on products. In 2014 one bank was ordered to pay $727 million in consumer relief for deceptive marketing and unfair billing of add-ons, and another to refund an estimated $309 million.9 In 2023 an auto lender was ordered to pay $60 million, including nearly $32 million to customers denied refunds of unearned GAP and credit insurance premiums, on products averaging $700 to $2,500 per loan.10
A quieter one, and it catches people who did everything else right.
Permanent policies let you borrow against the cash value, which is sold as access to your own money. Outstanding policy loans stood at about $135 billion in 2016, at fixed rates around 8% in early 2018 — against roughly 4.25% for a 30-year mortgage at the time.3
An unpaid loan compounds. If it consumes the cash value the policy can collapse, taking the death benefit with it and sometimes producing a tax bill on money never received.
Three questions. They work on every product in this chapter.
What are you paid on this, and how is it spread? On annuities you are entitled to a disclosure of the producer's cash compensation on request. Nobody volunteers it.
Can I compare this to an alternative on one number? If not, the difference between two versions of it is the part you are not being shown.
If this is a replacement, what does leaving the old one cost me? In cash and in tax, in writing, before anything is signed.
None of this is visible in a bank account. What is visible is the premium — the standing payment that continues for decades, unexamined, because it is small enough to disappear into a month. Plenee puts those payments beside everything else competing for the same money, and makes the annual question askable: is this still doing a job, and is it still the cheapest way to do it?
The issue is not that sellers are paid. It is that they are mostly paid in year one, which disconnects their reward from whether the product still suits you later. From that follow lapse-supported pricing, illustration complexity that survives a decade of regulatory patching, sales staged at the moment you cannot shop, and replacement that regenerates a commission at your expense. Home and auto insurance is the honorable exception and it is worth saying so. And the most portable rule in the subject: if you cannot compare two quotes on one number, the difference between them is the price you cannot see.
Plenee Academy provides financial information and education, not personalized financial advice. Plenee Co. is not a registered investment adviser, broker-dealer, or financial planner. Legal Disclosures & Notices →