Crypto Fund Insurance: What Crime, Specie and Directors Cover Actually Respond To

The useful question about crypto fund insurance is never whether cover exists. It is whose loss a policy is written to answer, and under which conditions. A custodian's own policy answers the custodian's loss, which is not the same as protecting the fund, and different again from protecting investors. Crime, specie and directors and officers cover respond to three distinct losses, and none substitutes for another. Managers and allocators routinely assume protection exists for events no policy class in this market is built around.

Managers are asked whether the assets are insured, and it is the wrong question, because the answer is almost always yes and almost always irrelevant. We ask instead whose loss the policy answers, and who holds the certificate.David Lloyd, Chief Executive Officer at CV5 Capital

Executive Summary

Neither the Mutual Funds Act (as amended) nor the Private Funds Act (as amended) imposes an insurance requirement on a Cayman fund. The requirement is commercial. The position stated here is current as at August 2026.

• Cover classes are defined by the loss they answer and the party they insure, not by the asset class marketed.
• Crime cover is built around dishonesty, principally insider acts and, by negotiated extension, social engineering.
• Specie cover attaches to property at rest, and its perimeter ends where trading begins.
• Directors and officers cover protects individuals against claims, never the portfolio against loss.
• A custodian's policy insures the custodian; the fund's recourse is contractual and capped.
• Exclusions and control warranties decide whether anything responds at all.

The Four Things a Digital Asset Fund Is Asked to Carry

Four distinct arrangements are compressed into the single word insured. The custodian's own cover, placed over its own operations. Crime cover, responding to dishonest taking of property. Specie cover, covering valuables held in a defined location. Directors and officers cover, responding to claims against individuals.

These are not tiers of one product. They sit in different parts of the market and answer different questions. Two errors recur in diligence: calling a custodian's arrangements the fund's cover, and calling directors and officers cover protection for assets.

None of it is required by Cayman fund legislation. The Directors Registration and Licensing Act requires directors of a CIMA-registered mutual fund to register with the Cayman Islands Monetary Authority, and imposes no cover requirement. Everything below is commercial practice or negotiated contract, not compliance.

Whose Loss Responds Is the Whole of Crypto Fund Insurance

Every question worth asking reduces to one test. If the event happens, who is the named insured, and whose loss does the policy indemnify?

A custodian places cover over its own vault and operations because it faces claims from clients whose property it holds. The fund is not usually a named insured and has no direct right of action against that insurer. What it holds is a contractual claim under the custody agreement, bounded by provisions that commonly cap recovery by reference to fees paid or a stated sum. The custodian's insurance sits behind that cap and does not enlarge it.

The certificate therefore matters more than the marketing. One naming the custodian as insured is a statement about that firm's balance sheet protection. It is useful when selecting a custodian and belongs in the fund's custody due diligence work, not in its answer on insurance.

Crime Cover: Dishonesty, Insiders and Social Engineering

Crime cover, sometimes written as a fidelity or commercial crime form, is a first-party class responding to the insured's own loss of property through defined dishonest acts. The organising idea is dishonesty. Error, negligence, market movement and technology failure are not what it is built for, however severe the loss.

Two triggers carry most of the weight. The first is the insider trigger, covering dishonest acts by an employee or, where extended, by personnel of the investment manager. Definitions of insured person and employee are negotiated and frequently narrower than assumed, particularly where trading, operations and technology sit in different entities.

The second is social engineering fraud, where an authorised person is deceived into transferring assets to an attacker. That trigger reflects how digital asset losses actually occur, and it is typically excluded from the base grant and added back by endorsement with its own sub-limit and conditions. Those conditions commonly require documented out-of-band verification of instructions and of address changes, and where the fund cannot evidence that step the endorsement may be of little use.

Where the insider definition bites

An insider acting within granted authority is the harder case. A signatory who exceeds a mandate or conceals losses may act improperly without meeting a definition requiring intent to obtain improper personal gain, and the remedy there is authority architecture, not a placement.

Specie Cover: Assets at Rest, and the Gap That Opens When They Move

Specie is an old first-party property class developed for valuables held in vaults. Applied to digital assets it attaches to private key material held in a described offline arrangement, in identified locations under identified controls. The subject matter is not the token on a public ledger but the key material and the arrangement holding it.

That construction has a consequence. Specie is built for property at rest, the described arrangement is the perimeter, and the grant is generally structured around loss from within it. Once key material is used in a signing environment connected to a network, or assets move to a venue, the property has left the arrangement the policy describes. Whatever protection exists then is a different question: a crime form, the venue's own arrangements, or nothing.

For a fund that trades, this is the gap that matters. Strategies hold working balances at venues and post collateral, and those balances sit outside a vault perimeter for as long as they are useful. A manager calling the fund insured on the strength of a specie arrangement over a cold reserve is describing protection for the part of the book least exposed to the strategy's risks.

Directors and Officers Cover: The Individuals, Not the Portfolio

Directors and officers cover is a third-party liability class. It responds to claims against individuals for alleged wrongful acts in their capacity, funding defence costs and, subject to terms, settlements and awards. It is not asset protection. No amount of it responds to the loss of portfolio assets, and presenting it otherwise is the commonest misdescription here.

The structure has three conventional parts: cover for individuals where the company cannot indemnify them, reimbursement where it has, and cover for claims against the entity. The mechanics appear in the treatment of directors and officers cover in a conventional fund, and the digital asset difference sits in the underwriting information rather than the architecture.

Two exclusions do most of the work. Conduct exclusions remove cover for deliberate dishonesty and improper personal profit, generally on final adjudication, which preserves defence costs until then. Insured versus insured exclusions restrict claims by one insured against another. Both matter to those who sit on the board, which is why cover terms arise when appointing independent directors on a digital asset fund, alongside the constitutional indemnity.

Exclusions and Conditions That Decide Whether Anything Responds

Cover classes are chosen once. Conditions and exclusions apply continuously, and they are where managers discover late that a placement does not meet the event in front of them. Nothing below describes a particular policy.

The protocol exploit line is the widest gap. First-party forms are organised around property being dishonestly taken. A contract behaving exactly as coded, an economic exploit of a lending market, or a governance vote redirecting value are not obviously a theft of insured property by a covered peril. On-chain exposure should be assumed to carry risks the market does not currently answer.

Self-Custody, Multi-Party Computation and Cover at Small Size

A fund holding its own key material is underwriting its own controls, and the market prices that accordingly. Multi-party computation changes the shape of the risk rather than removing it: there is no single key to steal, but signer set composition, quorum, recovery and vendor dependency all become underwriting information. Attestations and independently reviewed procedures make a described arrangement inspectable.

Size is the other constraint, and a market condition rather than a rule. Minimum premiums and retentions are set with larger programmes in mind, and control requirements do not scale down. A fund below institutional scale commonly finds its position is a custodian's contractual liability, a directors and officers placement, and control architecture doing the work a first-party policy is assumed to do. Saying so plainly is stronger than implying protection not placed.

Insurance therefore belongs in launch budgeting as a decision rather than an estimated line, interacting with custody model, venue count and board composition, all priced in the full launch cost stack.

What an Allocator's Operational Due Diligence Team Will Ask

Institutional review treats insurance as an evidence question. The reviewer is testing whether the manager understands the instrument, because a manager who misdescribes cover is likely to misdescribe other controls.

• Which entities are named insureds on each policy, and whether the fund appears anywhere.
• What the custody agreement liability cap is and what it excludes.
• Whether a specie limit over the custodian's arrangement is dedicated or shared on an aggregate basis.
• Whether social engineering sits inside the crime grant, and what it conditions.
• What operating model the cover was underwritten against, and what notification applies on change.
• Who reviews the position, how often, and where that review is recorded.

The last item is underweighted. A board that has considered the fund's protection, recorded its reasoning and identified the residual exposures presents better than one holding a larger limit it cannot explain. That review belongs with the operational due diligence readiness pack, and in the standing agenda in fund governance and operational due diligence readiness.

Key Takeaways

• Establish for every policy who the named insured is and whose loss is indemnified before calling the fund insured.
• Read the custody agreement liability clause and treat that cap as the fund's practical recourse.
• Confirm whether social engineering sits inside the crime grant, and evidence that its conditions are operated.
• Map which assets sit inside a specie perimeter and which sit at venues, and state that split.
• Manage on-chain protocol risk through limits, contract selection and authority architecture, not through cover.
• Minute a periodic insurance review, and notify the market whenever the custody or venue model changes.

Frequently Asked Questions

Is a Cayman digital asset fund required to carry insurance?

No. Neither the Mutual Funds Act (as amended) nor the Private Funds Act (as amended) imposes one, and the Directors Registration and Licensing Act addresses director registration rather than cover. The requirement is commercial and allocator-driven.

Does a custodian's insurance protect the fund's investors?

A custodian's policy is placed over its own operations and names the custodian as insured. The fund is not usually a named insured and has no direct right of action. Its recourse is the contractual claim under the custody agreement, bounded by the cap in it.

What is the difference between crime cover and specie cover?

Crime cover is organised around dishonest acts, principally insider dishonesty and, by endorsement, social engineering. Specie attaches to key material held in a described offline arrangement. Crime follows the act; specie follows the place.

Does directors and officers cover protect the portfolio?

No. It responds to claims against individuals for alleged wrongful acts in their capacity, funding defence costs and, subject to terms, settlements and awards. It does not respond to the loss of portfolio assets.

Would a smart contract exploit engage a fund's cover?

First-party forms are built around property being dishonestly taken from an insured party or location. Code executing as written, an economic exploit or a governance action do not sit comfortably there. Whether any wording responds is specific to that wording, so protocol risk is best managed through controls.

Can a fund below institutional scale place its own cover?

Availability is a market question and terms are negotiated individually, so no general answer holds. Minimum premiums and control requirements are calibrated to larger programmes, and smaller funds more often rely on contractual liability and a directors and officers placement.