Think of it like this: when an insurer offers a product where customers essentially own the underlying assets—like a with-profits policy—the insurer isn’t taking market risk. They’re collecting fees. So why should they book profits like they're betting on stocks? That’s the philosophical pivot IFRS 17 makes. The VFA flips traditional accounting logic on its head. It forces companies to treat themselves more like asset managers than risk-takers. And that’s where the real tension begins.
Understanding the Variable Fee Approach: Not Just a Model, But a Mindset Shift
Let’s start simple. The Variable Fee Approach isn't just another column in a spreadsheet. It’s a reflection of economic reality—when policyholders absorb investment volatility, insurers shouldn’t pretend they’re earning stable margins. Under VFA, the insurer's fee varies based on asset performance. If markets tank, profits drop. If they soar, so do fees. This is fundamentally different from the building-block approach or the premium allocation method. It’s performance-linked.
Contracts subject to VFA must meet strict criteria: direct participation (policyholders get a clear share of returns), significant discretion in dividends, and—this is key—exposure to underlying items that resemble investment contracts. We’re talking policies in places like Germany or Japan where bonuses are tied explicitly to fund performance. No vague promises. Real, traceable links.
But—and this is where people get tripped up—not all participating contracts fall under VFA. Some insurers tried to argue that “we’ve always paid bonuses based on profits,” but IFRS 17 demands transparency. If the return isn’t directly passed through, you don’t qualify. That said, meeting these conditions unlocks a unique profit recognition path. You estimate the expected cash flows, project returns on underlying items, and recognize profit as those returns crystallize. No front-loading. No smoothing illusions.
When Does VFA Apply? The Three Gatekeepers
Not every shiny insurance product gets thrown into the VFA bucket. There are three non-negotiable conditions. First: the contract must include a direct participating feature. That means policyholders aren’t just getting interest; they’re getting slices of actual portfolio gains. Second: the entity must have no significant risk or reward from changes in the underlying items. In other words, the insurer is a middleman, not an investor. Third: the amount returned to policyholders must vary with the performance of specific underlying items—like bonds or equities held by the insurer.
If any of these fail, you fall back to the general model. And that’s a big deal because VFA bypasses some of the more complex discounting mechanics. It’s almost… elegant. But only if you qualify. Many mutual insurers thought they were automatic candidates. They weren’t. Because even if they pass profits through, if the link isn’t explicit and measurable, VFA stays off-limits.
How Profit Emerges Under VFA: A Slow Drip, Not a Gush
Traditional insurance accounting let companies recognize profit early—sometimes immediately after sale. Not anymore. Under VFA, profit recognition is delayed until investment returns on the underlying items are realized. That means no more booking margin on day one. Instead, you wait. You watch. You adjust. As returns come in, a portion flows to shareholders as profit—the rest goes to policyholders.
The mechanism hinges on the fulfilment cash flows. You start with estimates of future cash inflows and outflows, including expected investment returns. Then, as time passes and actual returns emerge, you update your projections. The change in net cash flows due to investment performance feeds directly into profit or loss. But—and this is subtle—you only recognize the insurer’s share. The rest? It’s a liability to policyholders. So yes, you’re measuring gains, but only your cut counts as profit.
The Mechanics of VFA: Where Math Meets Judgment
Now let’s roll up the sleeves. Implementing VFA isn’t plug-and-play. It requires judgment calls masked as math problems. You have to identify the underlying items—say, a pool of corporate bonds backing a unit-linked fund. Then project their future returns using market-consistent assumptions. But what if some assets are illiquid? What if there’s no observable yield? That’s when models get messy. And that’s exactly where discrepancies between insurers begin to show.
The contractual service margin (CSM) still plays a role here, though differently than in the general model. It represents unearned profit. But under VFA, the CSM adjusts directly in response to investment return variances. Big return this year? CSM increases. Market crash? It might dip, unless offset by risk adjustments. But—and this trips up auditors—there’s no explicit risk adjustment required in the VFA. Why? Because the policyholder bears the risk. So the insurer doesn’t need a buffer for investment volatility. Only for non-financial risks like mortality or expense overruns.
I am convinced that this exclusion of investment risk adjustment is one of the most misunderstood parts of IFRS 17. People don’t think about this enough: removing that buffer makes earnings more volatile. But it also makes them more honest. You’re no longer hiding behind smoothed assumptions. What you see is what you get.
CSM Adjustments: The Pulse of VFA Profit Recognition
Each reporting period, the CSM is unlocked a little. The amount released reflects services provided during the period. But here’s the twist: when actual investment returns differ from expectations, the CSM gets adjusted. This isn’t a correction. It’s a reallocation. Higher returns mean more surplus to distribute—some to policyholders, some to shareholders. The release happens through profit or loss, which explains why some insurers saw earnings swings in 2023 when markets rebounded post-pandemic.
For example: A German life insurer using VFA reported a 14% increase in net income Q-over-Q in Q2 2023—not because sales spiked, but because bond yields rose and actual returns exceeded assumptions. Their CSM ballooned. That’s VFA in action. Not magic. Just mechanics working as intended.
Discount Rates and Assumptions: The Quiet Battleground
You can’t ignore discounting, even in VFA. While investment risk isn’t included in the discount rate (since it’s borne by policyholders), other risks still are. Mortality, lapses, expenses—you still apply a risk adjustment. But the rate itself must reflect the time value of money and liquidity characteristics of the liabilities. In high-inflation environments like 2022–2023, this became critical. Insurers using outdated yield curves faced massive restatements.
One European carrier had to revise its discount rates upward by 80 basis points after regulators challenged its use of pre-inflation crisis data. The result? A €210 million reduction in CSM. That’s not a rounding error. That’s boardroom news.
VFA vs. General Model: Which Fits Your Portfolio?
The general model is the default. VFA is the exception. Yet for the insurers that qualify, VFA offers cleaner alignment with economic substance. But it comes at a cost: increased earnings volatility. Under the general model, companies can smooth returns using locked-in rates. VFA doesn’t allow that. Your profits dance with the markets. In bull years, you look brilliant. In bear markets, not so much.
And that’s the trade-off. Do you want stability or transparency? Many CFOs quietly admit they’d prefer the old way. But regulators aren’t asking for stability—they’re demanding truth. As a result: insurers with significant participating business in markets like France or Switzerland are now splitting portfolios. One set under VFA, others under the general model. It’s not inefficient. It’s necessary.
General Model: Smoother but More Complex
The general model uses current discount rates and updates them every period. But it includes both financial and non-financial risks in the risk adjustment. It also allows more judgment in cash flow estimation. That flexibility can be a blessing—or a red flag for auditors. Unlike VFA, it doesn’t force profit recognition to track asset returns. So you can report steady margins even when markets wobble.
VFA: Transparent but Exposed
With VFA, there’s less room to hide. You can’t smooth. You can’t delay. Profit follows performance. That said, it reduces some complexity—no need to calculate separate risk adjustments for investment risk. But you better have clean data on underlying assets. Because if you can’t prove the link between fund returns and policyholder payouts, you lose VFA eligibility. And that changes everything.
Frequently Asked Questions
Can VFA Be Applied to Unit-Linked Products?
Only if they meet the direct participation criteria. Most unit-linked contracts don’t qualify because the insurer doesn’t hold the assets in a way that creates a direct link. The returns go straight to the policyholder without discretion. So no, not automatically. You need more than a fund connection—you need contractual discretion and participation mechanics that mirror traditional with-profits policies.
What Happens When Investment Returns Are Negative?
The CSM can go negative. But under IFRS 17, you don’t book a loss immediately. Instead, you reduce future profit recognition. Only if the deficit persists and expectations change do you recognize a loss. It’s a buffer, of sorts. But it’s not designed to last forever. If markets stay down, losses will eventually hit the income statement. No escaping gravity.
Is VFA Required for All Participating Contracts?
No. Only those where the policyholder bears significant investment risk. If the insurer guarantees returns or absorbs shortfalls, VFA doesn’t apply. Many U.S. participating policies, for instance, are backed by general account strength. They fall under the general model. It’s a bit like calling a scooter a motorcycle because it has two wheels. Same category? Maybe. Same engine? Not even close.
The Bottom Line: VFA Is Truth in Advertising for Insurance Profits
Let’s be clear about this: VFA isn’t about making accounting harder. It’s about making it honest. For decades, insurers booked profits on products where they were merely fee collectors. VFA ends that. It says, “If you’re not taking the risk, stop acting like you’re earning the upside.” That’s a philosophical shift disguised as a technical rule.
Some say it increases volatility. True. But volatility wasn’t invented by IFRS 17—it was just hidden before. Now it’s visible. And honestly, it is unclear whether markets will reward transparency or punish swings. Early data from 2023 shows mixed reactions. One insurer’s stock dropped 7% after a VFA-driven profit surge—investors suspected it wasn’t repeatable.
My take? I find this overrated. Yes, earnings bounce. But investors are smarter than we give them credit for. They can handle the truth. What they can’t stand is being misled. VFA strips away the fog. And in a world drowning in noise, that’s worth something. Suffice to say: if your business model relies on accounting opacity, you’ve already lost.