Internal:Training/IFRS17/The variable fee approach

🔗 Recall. In the previous page, you learned how the premium allocation approach simplifies measurement for short-duration contracts. Now we turn to the other end of the spectrum: long-duration contracts where the policyholder's returns are tied to underlying items, and where a different modification of the general model is needed to reflect their unique economics.

🎯 Objective. In this page, you will learn:

• What direct participating contracts are and why they behave differently from traditional insurance contracts.
• How the variable fee concept modifies the general model by treating changes in the insurer's share of underlying items as an adjustment to the CSM rather than sending them to profit or loss.
• What the three eligibility criteria are that a contract must satisfy before the variable fee approach can be applied.

~*~

What are direct participating contracts

🤝 A different kind of promise. Not all insurance contracts work like the traditional policies we have studied so far. In a traditional contract, the insurer promises a fixed or formulaic benefit: if your house burns down, you receive the repair cost; if you die during the coverage period, your family receives a set sum. But some contracts make a fundamentally different promise. They tell the policyholder: "We will invest a pool of underlying items on your behalf, you will receive a substantial share of the returns from those items, and we will also provide you with insurance coverage." These are direct participating contracts, and they sit at the heart of many life insurance and savings products sold across Europe.

💼 A familiar product. If you have encountered unit-linked life insurance or with-profits endowment policies, you have already met direct participating contracts in practice. Consider a with-profits savings product sold in France: the policyholder pays premiums that are invested in a pool of bonds, equities, and real estate. Each year, the policyholder receives a share of the investment returns generated by that pool, plus a guaranteed minimum. AXA, as the insurer, retains a fee for managing the assets and bearing the insurance risk. The policyholder's benefit is not fixed in advance; it depends directly on how the underlying items perform.

⚠️ Common misconception. Learners sometimes assume that any contract with an investment component qualifies as a direct participating contract. That is not the case. The defining feature is not merely the presence of an investment element, but the fact that the policyholder is promised a substantial share of the returns on a clearly identified pool of underlying items. A term life contract with a small cash-value rider, for example, would not typically qualify.

🔗 Why traditional accounting falls short. Under the general model, changes in financial assumptions, such as shifts in interest rates or equity prices, flow through to the income statement or OCI. For a traditional contract, that treatment makes sense: the insurer bears the financial risk. But for a direct participating contract, the insurer is essentially acting as an asset manager who also provides insurance coverage. When the value of underlying items rises, the policyholder's benefit rises too, and the insurer's share, its fee, adjusts accordingly. Forcing these financial movements into the income statement would create artificial volatility that does not reflect the real economics of the relationship.

🤔 Think about it. If the general model creates misleading volatility for these contracts, how might IFRS 17 redesign the mechanics to better reflect the insurer's true economics?

~*~

The variable fee concept: how VFA modifies the general model

💡 Reframing the insurer's role. The key insight behind the variable fee approach is that, for direct participating contracts, the insurer's profit is best understood as a variable fee charged for services rendered. That fee has two parts: a share of the returns on underlying items, and compensation for providing insurance coverage and investment management services. Instead of treating changes in underlying items as financial income or expense that hits the income statement, the VFA routes the insurer's share of those changes through the CSM. The result is a smoother, more representative profit pattern.

📊 How the mechanics differ. Recall that under the general model, changes in financial assumptions (such as discount rates or market returns) do not adjust the CSM; they flow to the income statement or OCI. The VFA changes this rule for direct participating contracts. Suppose AXA manages a with-profits portfolio in Belgium backed by a pool of underlying items worth €500 million. If equity markets rise and the pool's value increases by €25 million, the policyholders are entitled to, say, 90% of that gain (€22.5 million), and the insurer's share is 10% (€2.5 million). Under the VFA, that €2.5 million increase in the insurer's share adjusts the CSM upward. It will be released as insurance revenue over future periods as the insurer continues to provide service.

⚠️ Common misconception. A common error is believing that the VFA eliminates all volatility from the income statement. It does not. Changes in non-financial assumptions, such as revised mortality or lapse expectations, are treated the same way as under the general model: changes relating to future service still adjust the CSM, but changes relating to current or past service still go to profit or loss. What the VFA specifically adds is the routing of financial changes through the CSM, something the general model does not permit.

🔎 Seeing it in context. To appreciate why this matters, imagine the Belgian portfolio at year-end after a sharp stock market decline. The pool's value drops by €30 million. Policyholders absorb €27 million (90%), and the insurer's share falls by €3 million. Under the general model, that €3 million would hit the income statement immediately, creating a visible loss even though the insurer's long-term fee structure is unchanged. Under the VFA, the €3 million reduces the CSM instead. As long as the CSM remains positive, the insurer simply recognises less profit in future periods, matching the economics of a fee that fluctuates with the assets it manages.

🤔 Think about it. The VFA clearly suits contracts where policyholders share in investment returns. But not every contract with some participation feature qualifies. How does IFRS 17 draw the line between contracts that may use the VFA and those that must stay under the general model?

~*~

Scope and the three eligibility criteria

📏 A strict gateway. IFRS 17 does not allow insurers to choose the variable fee approach freely. A contract must satisfy three criteria at inception, and all three must be met simultaneously. If even one fails, the contract must be measured under the general model (or the PAA if eligible). The three criteria are designed to ensure that the contract genuinely represents a fee-for-service arrangement tied to underlying items, rather than a traditional contract with a thin investment wrapper.

1️⃣ Criterion one: a specified pool of underlying items. The contractual terms must specify that the policyholder participates in the returns of a clearly identified pool of underlying items. These items might be a portfolio of bonds, an equity fund, a basket of real estate holdings, or any other identifiable collection of assets, liabilities, or a combination. The pool must be identifiable; a vague reference to "the insurer's general investment portfolio" would not qualify. For example, a unit-linked product in Italy where each policyholder's account is linked to a named fund of European equities clearly satisfies this criterion.

2️⃣ Criterion two: a substantial share of fair value returns. The insurer must expect to pay the policyholder a substantial share of the fair value returns on those underlying items. "Substantial" is not defined as a specific percentage, but the standard requires that the policyholder's share be significant enough that the contract is genuinely participatory. A product that promises the policyholder 90% of the returns on an identified bond fund easily meets this test. A product that promises only 5% of returns, with 95% retained by the insurer, would not, because the policyholder's participation is not substantial.

⚠️ Common misconception. Some learners interpret "substantial share" as meaning the policyholder must receive the majority (over 50%) of the returns. The standard does not set a fixed threshold. What matters is the overall economic substance: the policyholder's share must be large enough that the contract's cash flows are expected to vary substantially with the performance of the underlying items. The assessment is made at inception and considers the full range of scenarios, not just the most likely outcome.

3️⃣ Criterion three: substantial variability with underlying items. The insurer must expect that a substantial proportion of any change in the amounts paid to the policyholder will vary with the change in fair value of the underlying items. This criterion zooms in on the variability of the total benefit, not just the existence of participation. Consider a product that offers both a guaranteed floor and a participation feature. If the guarantee is so generous that the policyholder will almost always receive the guaranteed amount regardless of how the underlying items perform, then the benefit does not substantially vary with those items, and criterion three fails. The contract might look participatory, but the guarantee dominates the economics, making the general model more appropriate.

🧩 Putting the pieces together. All three criteria work as a single gate. A with-profits savings product at an insurer like AXA that names a specific asset pool, promises the policyholder 85% of the returns, and has only a modest minimum guarantee would typically pass all three tests and qualify for the VFA. Conversely, a guaranteed annuity product that happens to hold an identifiable asset pool but promises the policyholder a fixed benefit regardless of performance would fail criteria two and three, even though criterion one is met. The three criteria ensure that the VFA is reserved for contracts where the insurer truly acts as a fee-collecting service provider, not a traditional risk bearer.

~*~

Takeaways

📌 Key takeaways.

• Direct participating contracts promise the policyholder a substantial share of the returns on a specified pool of underlying items, making the insurer's profit function like a variable fee for service.
• The VFA modifies the general model by routing the insurer's share of changes in underlying items through the CSM, producing a profit pattern that reflects the economics of a fee-based relationship rather than creating artificial volatility.
• A contract qualifies for the VFA only if it meets all three eligibility criteria at inception: a specified pool of underlying items, a substantial policyholder share of returns, and substantial variability of benefits with those items.

~*~