An examination of how legal substance is evaluated when form alone is insufficient
By Ahmad T. Sulaiman
Principal, Atlas Citadel Group
The legitimacy of a captive insurance company is rarely decided at the moment it is formed. Licensure, documentation, and elections establish a starting point, but they do not resolve the question that ultimately matters. That question is whether the captive functions as insurance over time.
For regulators, courts, and tax authorities, the distinction between real insurance and arrangements that merely resemble it is not philosophical. It is practical. The analysis does not turn on intention, marketing language, or structural complexity. It turns on behavior. Specifically, it turns on whether the arrangement satisfies a small number of enduring tests that have defined insurance for decades.
These tests are neither novel nor obscure. They are applied repeatedly across regulatory examinations and judicial decisions. They are straightforward in concept, yet exacting in application. A captive that satisfies them consistently is treated as insurance. A captive that does not eventually reveals the gap between form and substance.
The first test is whether risk has genuinely transferred from the insured to the insurer.
Risk transfer is often assumed to occur when a policy is issued or a premium is paid. In practice, it occurs only when the insurer bears the financial consequence of loss. The decisive question is simple. When a covered event occurs, who pays.
If the operating business absorbs the loss regardless of policy language, risk has not transferred. If the captive lacks the capital or legal obligation to respond, risk has not transferred. If side arrangements or informal understandings ensure that the insured is effectively reimbursed outside the policy, risk has not transferred.
Risk transfer becomes real only when the captive pays claims from its own assets under binding terms, without regard to ownership or convenience. Until that point, the arrangement remains theoretical.
The second test is whether risk is distributed.
Insurance does not operate by predicting individual outcomes. It operates by spreading uncertainty across multiple exposures so that losses become manageable in the aggregate. Without distribution, insurance collapses into speculation on isolated events.
A captive insuring a single exposure or a single outcome struggles to meet this test. Distribution is typically achieved by insuring multiple operating entities, multiple lines of coverage, or by participating in pools of unrelated risk. What matters is not the elegance of the structure but the effect. Losses must move across participants in a manner that prevents any single event from determining the financial fate of the insurer.
Regulators do not look for formal symmetry. They look for evidence that risk is shared rather than recycled.
The third test is whether the risk being insured is fortuitous.
Insurance covers events that may or may not occur. It does not cover costs that are inevitable, scheduled, or entirely within the control of the insured. This distinction is often misunderstood because many business expenses feel uncertain in practice even when they are predictable in principle.
A lawsuit may be filed or it may not. A cyber incident may occur or it may not. A regulatory action may arise or it may not. These are fortuitous risks. Routine payroll, planned capital expenditures, and ordinary maintenance are not.
When a captive begins to insure predictable costs rather than uncertain events, it ceases to function as insurance and begins to resemble a budgeting mechanism. That shift is subtle, but it is determinative.
The fourth test is whether the captive operates like an insurance company.
This test is often the most revealing because it focuses on conduct rather than structure. Real insurance companies underwrite risks before issuing policies. They price premiums using actuarial methods rather than statutory targets. They establish reserves and revise them as information develops. They administer claims through defined procedures. They document decisions contemporaneously. They hold board meetings that reflect oversight rather than formality. They invest conservatively, prioritizing liquidity and claims paying ability over yield.
Individually, these practices appear unremarkable. Collectively, they are decisive.
A captive that performs these functions consistently over time appears ordinary under examination. A captive that performs them selectively or retrospectively does not.
What gives these tests their force is the way they reinforce one another. Risk transfer without distribution is fragile. Distribution without fortuity is artificial. Fortuity without operational discipline is theoretical. Operations without claims are hollow.
When all four elements are present, insurance emerges naturally. When any one is missing, the structure eventually reveals itself, often long after formation, when habits have replaced intentions.
This explains why scrutiny so often surprises owners. Captives rarely fail at inception. They fail gradually, through accumulated silence, deferred claims, and informal practices that feel harmless in isolation. By the time examination occurs, the issue is no longer how the captive was described. It is how it behaved.
The most defensible captives are not those that are most complex or innovative. They are those that are most ordinary. They behave consistently. They accept loss as part of operation. They document decisions as they are made. Over time, that consistency becomes substance.
The four tests are not hurdles to be cleared once. They are conditions to be maintained continuously.
A captive that satisfies them quietly does not need to argue that it is insurance. It demonstrates that fact through its conduct.
Author Information
Ahmad T. Sulaiman, Esq., is Principal of Atlas Citadel Group and Managing Partner of Atlas Law Center, a division of Sulaiman Law Group, Ltd., a national consumer protection and labor and employment law firm based in Chicago, Illinois. He completed his legal studies at Loyola University Chicago School of Law and Harvard Law School, and in 2026 completed the Chief Artificial Intelligence Officer program at the University of Chicago Booth School of Business as part of its inaugural class. He is a published author on banking law and consumer rights, including Consumer Defense: The Luxury of the Informed. His work focuses on extending institutional-grade tax, insurance, and risk architecture to sectors historically excluded from such tools, with emphasis on aligning insurance economics, federal tax law, and regulatory substance. The views expressed are his own.
