Captive Insurance for Business Owners: §831(b), Legitimate Structures, and IRS Scrutiny (2026)

A captive insurance company is a licensed insurer that a business creates to cover its own risks. Done correctly, it can fund legitimate business coverage while generating significant tax benefits. Done wrong — or with an aggressive promoter — it lands you on the IRS "dirty dozen" list with penalties and back taxes. The difference matters enormously, and most business owners don't know which side of that line they're on.

What is a captive insurance company?

A captive is a genuine insurance company — licensed in a state (Vermont, Delaware, Hawaii, Utah) or offshore (Cayman Islands, Bermuda, Barbados) — that your business forms and controls. Instead of paying premiums to a third-party insurer for coverage you may never use, you pay premiums to your own captive. The captive holds reserves, invests them, and pays claims when losses occur.

From a risk management perspective, captives make sense when:

• You're paying high premiums for coverage in the commercial market that doesn't exactly fit your risks
• Your loss history is better than industry average, but you can't get pricing credit from commercial insurers
• You have identifiable risks that are hard or expensive to insure commercially (supply chain interruption, product recall, reputational risk, cyber gaps)
• You want reserves to accumulate and grow over time under your control

The tax advantage is layered on top of the risk management rationale — not the other way around. That ordering matters, because the IRS scrutinizes structures where the tax benefit appears to be the primary purpose.

The §831(b) election: how it works

Under §831(a), insurance companies pay federal income tax on their net underwriting income (premiums minus claims, reserves, and expenses). A small insurance company with a good loss ratio would pay meaningful tax on its underwriting profit.

§831(b) offers an alternative: if a non-life insurance company elects and its net written premiums do not exceed the annual threshold, it pays tax only on its investment income — not on premiums received. The premium income is excluded from gross income entirely.¹

2026 §831(b) premium limit

For taxable years beginning in 2026, the §831(b) premium limit is $2,900,000 — up from $2,850,000 in 2025. This limit is indexed annually for inflation and rises in $50,000 increments (IRS Rev. Proc. 2025-32).¹

Tax math on a $831(b) captive

Suppose a manufacturing business with $8M revenue pays $600,000/yr in premiums to its captive, covering risks the commercial market doesn't price well (equipment breakdown, supply chain, product liability gaps). The captive makes a §831(b) election:

• The operating company deducts $600,000 in insurance premiums as a business expense (§162)
• The captive receives $600,000 in premiums — all excluded from gross income under §831(b)
• The captive earns $30,000 in investment income on reserves — taxed at 21% corporate rate ($6,300)
• Net tax result: $600,000 deduction at 37% (owner's combined rate) = $222,000 tax savings; tax cost at captive level ≈ $6,300

When the owner eventually liquidates the captive or distributes accumulated reserves, those distributions are taxed — typically at 20% capital gains + 3.8% NIIT = 23.8% federal. The deduction-to-capital-gains arbitrage (37% in → 23.8% out) is the core economic benefit when the captive is run legitimately over time.

What makes a captive legitimate

1. Risk shifting

The insured must genuinely transfer economic risk to the captive. If a loss occurs, the captive pays — not the parent company out of its own funds. A captive that is undercapitalized, relies on a parental guarantee, or returns all losses back to the parent economically has not achieved real risk shifting.

2. Risk distribution

Risk distribution requires that the captive insure enough independent risks that the law of large numbers operates — no single claim can wipe out the captive. In the single-parent context, courts have found risk distribution can be achieved by: (a) insuring multiple entities in a group, (b) entering a risk-pooling arrangement with unrelated captives, or (c) writing third-party coverage. A captive that insures only one entity and one type of risk is vulnerable on distribution grounds.

3. Arm's-length premiums

Premiums must be actuarially justified — set by a qualified actuary based on the actual risk profile, historical loss data, and commercial market comparisons. Premiums inflated beyond what an arm's-length insurer would charge are the most common IRS challenge point. A premium study from a credentialed actuary (FCAS or ACAS) is non-negotiable for a defensible structure.

4. Genuine claims payment

The captive must actually pay claims when losses occur. Structures where the captive "insures" risks that never materialize — and where premiums simply accumulate with no real loss history — look abusive because they are. Captives should cover real risks with real loss potential.

5. Separate governance

The captive must be a genuinely separate entity with independent governance, its own board, separate bank accounts, proper capitalization, and annual actuarial reviews. Captives run as shells with no operational independence fail the substance test.

The IRS scrutiny problem

Since 2016, the IRS has treated many §831(b) microcaptive arrangements as "transactions of interest" — a category of potential tax shelters requiring disclosure and attracting IRS examination. The problem grew from promoters selling cookie-cutter captive structures that were designed primarily to generate premium deductions with little genuine insurance purpose.

Notice 2016-66 and its aftermath

IRS Notice 2016-66 identified §831(b) microcaptive transactions as transactions of interest when the captive exhibits one or more abusive hallmarks: premiums that significantly exceed comparable commercial coverage, coverage for unlikely or implausible risks, a loss ratio below 70% over a representative period, or the use of reinsurance arrangements that funnel value back to the insured's owner rather than independent third parties.³

Courts have largely upheld the IRS's authority to challenge abusive microcaptives. The IRS has won a substantial majority of microcaptive Tax Court cases, obtaining accuracy-related penalties (20%) and, in egregious cases, civil fraud penalties (75%) in addition to back taxes and interest.

2025-2026 reporting requirements

Notice 2025-24 updated the disclosure requirements for microcaptive reportable transactions. Participants in reportable microcaptive transactions must file Form 8886 disclosures with the Office of Tax Shelter Analysis (OTSA). Material advisors (promoters and attorneys who structured the arrangement) face their own Form 8918 filing obligations. Failure to disclose a reportable transaction can result in penalties of $10,000–$100,000 per year, even if the underlying arrangement is not ultimately disallowed.⁴

IRS settlement offers

The IRS has made settlement offers to certain groups of microcaptive participants under audit, generally requiring repayment of a portion of claimed deductions plus interest, with reduced penalties for participants who cooperate. If you are in an existing captive arrangement that was promoted with aggressive tax benefits as the primary focus, consulting a tax attorney before responding to any IRS correspondence is critical.

Abusive vs. legitimate: the practical distinction

Who captive insurance makes sense for

Captive insurance — even a legitimate, defensible structure — carries significant complexity and cost. It makes sense only when the risk management benefit and the economic scale justify those costs. General criteria:

• Revenue: Typically $5M+ in annual revenue. Below that, the economics rarely work — setup and compliance costs eat the benefit.
• Real coverage gaps: You have specific, quantifiable risks that the commercial market either can't cover, prices unfairly, or prices based on industry averages that don't reflect your actual loss history.
• Cash flow to fund premiums: Premiums into the captive are real cash out of the business. The captive holds reserves; they're not immediately accessible. You must be able to tolerate reduced liquidity.
• Long-term horizon: Captives are 10-20+ year programs. The deduction-to-capital-gains arbitrage takes years to play out. Businesses within a few years of sale should evaluate carefully — liquidating a captive before long-term capital gains rates apply eliminates the arbitrage.
• Willingness to operate it as a real company: Annual actuarial reviews, licensed management company fees ($25K-$75K/yr), state filing fees, board meetings, claims processes — this is not a passive structure.

Setup and ongoing costs

A domestic captive (Vermont, Delaware, Hawaii, or other licensed jurisdictions) typically costs:

• Setup: $30,000–$75,000 (feasibility study, domicile selection, actuarial analysis, legal documentation, licensing fees)
• Annual management: $25,000–$60,000 (captive management company fees, annual actuarial review, state premium taxes, regulatory filings, audit)
• Annual actuarial: $10,000–$25,000 depending on complexity
• Total annual cost: $40,000–$100,000+ before considering the premiums themselves

For a captive generating a $600,000/yr premium deduction at a 37% combined rate, the gross tax benefit is roughly $222,000/yr. After annual management costs of $70,000, net benefit is approximately $152,000/yr — before accounting for the eventual capital gains tax when reserves are distributed. This math works for genuinely well-run programs at scale; it fails quickly for under-$200K/yr premium programs.

The specialist advisor's role

Captive insurance sits at the intersection of insurance regulation, tax law, actuarial science, and corporate governance. A generalist financial advisor won't know how to structure it; a generalist CPA won't know the insurance law; and a captive promoter has a conflict of interest in recommending it. A fee-only financial advisor who specializes in business owner planning can help you evaluate whether a captive makes sense relative to your other options, and can connect you with qualified captive counsel and actuaries rather than promoters who earn commissions on setup.

If you have an existing captive arrangement that was set up by a promoter and haven't recently reviewed it against the IRS scrutiny criteria above, a specialist review is worth doing before your next filing season.

Summary: the honest answer

A well-structured, legitimately operated captive insurance company is a real and defensible tax strategy for the right business. The §831(b) election can create meaningful deduction-to-capital-gains arbitrage over a 10-20 year program. The catch: it only works if the captive is genuine insurance — actuarially priced, appropriately capitalized, paying real claims, and run as an independent entity. The IRS has made clear it will pursue arrangements that don't meet that standard, and the penalties for getting it wrong are severe. For most business owners under $5M revenue, simpler strategies (retirement plan stacking, QSBS planning, bonus depreciation) offer comparable tax value with far less complexity and IRS exposure. For larger operators with real coverage gaps and the resources to run it properly, captive insurance deserves a serious look.