Health insurance: basic cover, smarter bet?
Adverse selection, instability, repricing: which health insurance product is actually the more responsible one?
For non-French readers: French health insurance regulations distinguish between "regulated" contracts — which must cover a defined basket of care with no out-of-pocket costs — and "non-regulated" contracts, which sit outside that framework. This article explores why the latter may, paradoxically, be the more responsible choice.
At Seyna, our first health insurance programmes launched at a loss ratio of 150%. For those who don’t live with that number day-to-day: it means we were paying out €1.50 in claims for every €1 of premium collected. At that level, you never recover. It’s essentially definitive.
And after a blow like that, the reflex that follows is often worse than the original mistake. You lock everything down, stack margin upon margin. You rebuild the pricing in bunker mode. To make sure it never happens again. A few months later, you discover you’ve built a product nobody wants to buy. Production close to zero. Teams spinning their wheels. Losses that are different, but just as real.
These two mistakes don’t cancel each other out. They follow one from the other. Always in the same order. And they have the same root cause: a flawed understanding of where the real risk in health insurance actually comes from.
The “100% Health” reform covered French patients well. It also built an adverse selection machine.
Since 2020, premium increases in individual health insurance have been almost permanent. Double digits, every year, across virtually the entire market. This is not a coincidence. It is not the fault of any one player. It is the logical consequence of a system that created rational incentives for everyone — except the insurer.
Here is what happens in practice. A policyholder finds out they need five crowns replaced. Quote: €3,000. They take out a “regulated” health policy, get reimbursed, then cancel. They may have paid €600 in premiums over the year. The transaction is perfectly rational from their perspective. From the insurer’s, it’s a loss ratio of 500% on that single contract.
And it’s not just a matter of policyholder behaviour. Healthcare providers have also adapted their practices to a system that guarantees full reimbursement. Billing at the ceiling, add-on procedures, premium equipment systematically offered. Nobody is cheating. Everyone is rationally optimising within a system that allows it. And it is the insurer who absorbs the cost.
What this produces at portfolio level: explosive claims in year one, mass lapses once the care has been received, and a permanent race between premium increases and the departure of those who use their cover the least. Managing in those conditions is like walking a tightrope over a void, with the broker watching.
Non-regulated cover is not a cut-price version. It is a different contract, for a different policyholder.
By stepping outside the “100% Health” basket, the non-regulated contract does one simple thing: it removes the main incentive to take out a policy for a one-off need that has already been identified. And in doing so, it radically changes the profile of who comes knocking.
In a regulated contract, a significant portion of policyholders arrive with a care plan already mapped out. In a non-regulated contract, the policyholder comes because they want cover. For routine consultations, unexpected events, a hospitalisation they have not already scheduled for three months from now. It is not the same risk profile, not the same intent, nor the same lapse dynamic.
Our thesis — which the data is beginning to confirm — is that this policyholder has a structurally lower tendency to lapse. They did not come to optimise a dental procedure; their initial need is to be covered. Waiting periods reinforce this effect further: they filter out opportunistic sign-ups with a very short time horizon. The result is more stable portfolios, more readable claims experience, and day-to-day management that does not feel like permanently defusing a bomb.
But to be clear: the regulated contract does not deserve to be discarded. For someone with genuine recurring needs — significant dental work, real hearing loss, a heavy optical prescription — it is the right product. These are two different logics, two different policyholder profiles. A market that only offered one of the two would be cutting itself off from a significant share of its legitimate customers.
The problem is not the regulated contract. It is that it creates an opportunity that the most savvy policyholders seize — and one cannot blame them for it.
The non-regulated contract does, however, have one condition for effectiveness that the market underestimates.
Since 2020, French policyholders have grown used to full reimbursement on dental, optical and hearing care. That is their frame of reference. A policyholder who takes out a non-regulated contract without understanding they are stepping outside that framework will discover the out-of-pocket costs at the worst possible moment — when they actually need care. It is not the product that is at fault; it is the lack of clear explanation at the point of sale.
A poorly explained non-regulated contract is a ticking time bomb: complaints, lapses, and reputational damage. It is not the product that creates disappointment. It is a promise that was miscalibrated at the outset.
The non-regulated contract requires more client education at the point of sale, not less. It is a real insurance contract, covering real risks, but one that asks the policyholder to understand what they are buying — and the broker to play their proper role.
Whatever the contract, year one is not the year of reckoning.
This holds true in the regulated segment, where no insurer genuinely wins in year one, and where everything hinges on the ability to retain the right block and reprice without losing the portfolio. It also holds true in the non-regulated segment, even if the trajectory is more stable and more readable.
What differentiates insurers on this point is not their ability to build a good initial price. Anyone can put together a competitive year-one premium. It’s actually quite easy — and potentially dangerous. We learned that the hard way.
The dominant market approach: launch, observe, send a letter at year end. “+18% from 1 January.” The broker discovers they must call their clients to announce an increase they never saw coming, on a product they sold on the promise of stability. A bad moment for everyone — for the client who does not understand, for the broker who loses confidence, for the insurer who loses a partner.
Our approach is different. At the point of commercial proposal, we lay out the rules of the game: here is the premium, here is why it is competitive in year one, and here is what we will probably need to do on the in-force book in year two to stay on a balanced trajectory. The repricing is no longer a surprise — it is a strategic conversation held before launch.
A broker who knows how things will unfold can anticipate, prepare their clients, and defend the increase with real arguments. It is a different kind of relationship — more robust, and paradoxically, more commercial.
So, "non-responsible"... really?
Technically, yes — it sits outside the “100% Health” framework. But from an insurance standpoint, it is often the more responsible contract. It covers a real risk, attracts policyholders who are seeking lasting protection rather than a one-off optimisation, and produces portfolios that are more stable and more manageable.
What is less responsible, on the other hand, is selling any contract without being honest about what it covers, who it is designed for, and how it will evolve over time.
A contract does not become responsible because it carries that name. It becomes responsible when it remains insurable over time.
