Why self-funding has moved up the agenda
For a large Indian employer, group medical and group personal accident cover has stopped behaving like a fixed cost. Medical inflation runs ahead of general inflation, claims ratios on corporate health books have hardened, and each renewal arrives with a loading that the fully insured model passes straight through to the employer. The buyer carries the cost of the risk but captures little of the upside when claims come in better than priced.
Self-funding inverts that. Instead of paying a premium that bundles expected claims, the insurer's margin and a risk load, the employer funds claims as they fall and keeps whatever it does not spend. The catch is volatility: a self-funder is exposed to the bad year as well as the good one, and a single catastrophic claim or a worse-than-expected aggregate can turn a saving into a shock.
The risk-financing answer is to self-fund the predictable layer and insure the tail. An employee-benefit captive applies a self-insurance structure to healthcare, letting the employer self-fund while combining that with stop-loss insurance that caps catastrophic claims. The captive absorbs the mid-sized claim volatility it can comfortably carry, and the stop-loss protects it against the large, unexpected losses it cannot. That division of labour, retain the frequency, transfer the severity, is the heart of the playbook.
The two stop-loss layers that make self-funding safe
Self-funded medical programmes are protected by two distinct stop-loss layers, and most self-funded employers carry both because each guards against a different failure mode.
Specific stop-loss
Specific stop-loss caps the claims attributable to any single covered member. Once one individual's claims in the policy year cross the specific attachment point, the stop-loss insurer takes the excess. This is the protection against the catastrophic individual case, the long ICU admission, the high-cost oncology or transplant pathway, that would otherwise blow a hole in a single year's fund.
Aggregate stop-loss
Aggregate stop-loss caps the total claims across the whole covered population. It responds when the sum of all claims, none of which need individually breach the specific point, exceeds an annual aggregate attachment, usually set as a percentage above expected claims. This is the protection against a bad year in the round, a flu season, a cluster of moderate claims, a general deterioration in experience.
The attachment points on these two layers are the dials that set how much risk the captive actually retains. Setting them too low transfers most of the volatility back to an insurer and erases the saving; setting them too high leaves the captive exposed to swings it is not capitalised for. Calibrating them against the group's own claims data is the central actuarial task.
Where the captive and the fronting structure fit
A captive is the vehicle that holds the retained risk. For an Indian group, the captive may sit in an offshore domicile or, increasingly, in the GIFT City IFSC, and it works alongside a fronting insurer that issues the policy the employees actually hold while the captive reinsures the agreed retention behind it.
The fronting carrier matters for practical reasons. Employees and local regulators deal with an admitted, regulated insurer; the captive sits behind that paper carrying the economic risk. The fronting insurer also issues the local policy, handles the regulated administration, and cedes the retained layer to the captive, with the stop-loss attaching above the captive's retention.
This is not a quick structure to stand up. Captive-funded multinational employee-benefit programmes typically begin through a multinational pooling process and require constant engagement with underwriters, actuaries, fronting carriers, local regulators and policy administrators, an undertaking measured in years rather than months. The pooling stage lets a group consolidate the experience of its various local benefit policies, see the true loss picture, and then decide how much of it the captive should take. Skipping that groundwork in favour of an immediate captive retention is how programmes get mispriced.
The governance load is the real entry barrier. A captive is a licensed insurer with capital, reserving, audit and reporting obligations, and an employee-benefit captive adds the actuarial discipline of pricing health risk and the regulatory complexity of multiple jurisdictions on top.
Pooling brings self-funding below the largest corporates
The assumption that self-funding employee benefits is only for the very largest employers is out of date. Group and employee-benefit captives are being used by employers in roughly the 50 to 1,500 employee range to pool stop-loss risk with like-minded employers, which makes self-funding a safer path well below the scale at which a single employer could carry the volatility alone.
The mechanism is pooling. A mid-sized employer's own population is too small for the law of large numbers to smooth its claims; one or two catastrophic cases can dominate a year. By pooling stop-loss risk with other comparable employers, often through a group captive or cell structure, the participants share the volatility across a much larger combined population, and the experience becomes stable enough to fund predictably. Each member still self-funds its own predictable layer, but the catastrophic and aggregate tail is diversified across the pool.
This changes the strategic question for a mid-sized Indian employer or a GIFT City based capability centre. The decision is no longer whether the organisation is large enough to self-insure on its own, but whether it can join or form a pool that gives it access to the economics of self-funding with the stability of shared risk. A protected-cell facility or a group captive can offer that on-ramp without the full cost of a standalone captive.
Sequencing the move and the governance it demands
A move from fully insured group benefits to a captive-backed self-funded programme is a multi-year transition, and treating it as one avoids the two common failures: under-capitalising the retention, and under-resourcing the governance.
A disciplined sequence runs roughly as follows.
- Pool and gather data first. Consolidate the group's benefit experience through a multinational pooling process so the true loss picture is visible before any retention is set.
- Model the retention. Use the group's own claims data to calibrate the specific and aggregate stop-loss attachment points, choosing how much volatility the captive should carry.
- Choose the vehicle. Decide between a standalone captive, a cell, or membership of a group captive, and choose the domicile, with GIFT City IFSC now a live option for Indian groups.
- Appoint the fronting carrier and administrators. Put in place the admitted insurer that issues local policies and cedes to the captive, and the administration that runs claims.
- Build the ongoing governance. Stand up the actuarial reserving, audit, capital management and regulatory reporting the captive needs as a licensed insurer.
The payoff is a risk-financing structure where the employer keeps the savings on a good year, carries only the volatility it is capitalised for, and transfers the catastrophic tail it cannot. The cost is genuine: capital, multi-year setup, and continuous engagement with actuaries, underwriters, fronting carriers and regulators. For a group with the scale or the pool to support it, that trade is increasingly worth making as fully insured costs harden.
Getting the retention, fronting and stop-loss layers right depends on reading how the underlying group-health and stop-loss wordings actually operate, where the attachment points bite, how the fronting cession is documented, and how the catastrophic layer is triggered. Sarvada gives commercial insurance brokers structured, searchable access to insurer policy wordings and the intelligence around them, so an employee-benefit captive is built on the specific contract terms behind it. Request Access to ground your risk-financing strategy in the wording detail.
