Outsourcing to Yourself

Moody’s has put a figure on something the life insurance industry would rather discuss in the passive voice. The fees that US insurers pay their own affiliates to manage the money inside offshore reinsurance structures have grown by around forty per cent since 2015.¹ The agency files this, reasonably enough, under counterparty risk and transparency. But the number is more interesting than the category it has been placed in. It is not a measure of a cost the industry has been forced to bear. It is a measure of how well the structure has been working.

Begin with the word “outsourcing,” which is doing a great deal of quiet work. Outsourcing implies distance: you hand a function to someone else because they can do it better or cheaper, and the fee you pay them is the price of that arm’s-length advantage. That is not what is happening here. The insurer cedes its liabilities to a reinsurer it is affiliated with, frequently sitting in Bermuda or the Cayman Islands. The assets backing those liabilities are then managed by an asset manager the insurer is also affiliated with. The fee for that management flows up to the same parent that owns both ends of the trade. The insurer has not sent the work outside. It has sent it across the room and billed itself for the journey.

This is the part the headline obscures. A forty per cent rise in fees, in an ordinary commercial relationship, is a problem to be managed — a margin leaking out of the business toward an external counterparty. In an affiliated structure it is the opposite. The fee is not leaking out; it is the channel through which value is deliberately routed to where the owner wants it. The regulated insurer is the raw material. The annuities and the long-dated liabilities they sit against are the feedstock. What the structure actually produces, and what its owners are actually in the business of producing, is fee-earning assets under management. The insurance is the wrapper. The fee is the product.

Once you see it that way, the trajectory of the figure stops being surprising. The owners of these structures are, overwhelmingly, private capital — alternative asset managers and the firms that have spent the past decade buying life and annuity books precisely because those books come with a permanent, captive, slow-moving pool of money to manage. A life insurer’s liabilities run for thirty years. The fee on the assets behind them runs for thirty years too. There is no better annuity, for an asset manager, than an actual annuity. The forty per cent is not drift. It is the design maturing.

The shape of the bill, where any of it can be seen, bears the point out. The investment-management agreements the offshore reinsurers strike with their affiliated managers — read through now in some number by the trade press — run to a dizzying menagerie of charges: flat fees and performance fees, fees for advice and for reporting, fees for parting, and, in at least one agreement, a clause guaranteeing the manager a set percentage of the assets or, should that fall short, a stated minimum in dollars whatever the year has done. A fee with a floor beneath it is a strange sort of price; it is the price a party sets when it is sitting on both sides of the table. And the parties collecting these fees are, with some regularity, the same people, or the colleagues of the same people, who sit on the boards of the insurers paying them.¹

*  *  *

None of this is new. The clearest precedent for it sits in a market that no one in London or New York was paying attention to: the South African medical schemes of the late 1990s. It is worth dwelling on, because it is the same structure, diagnosed to the bone, a quarter of a century early.

In May 2000 the Council for Medical Schemes met to consider a report it had commissioned into how the country’s medical schemes were using reinsurance. The schemes were small, the market parochial, the sums modest by the standards of global finance. The report nonetheless reached a conclusion that ought to have travelled further than it did. In the words of the actuaries who wrote it up the following year, reinsurance had become “a conduit for systematically removing surplus” from the schemes it was meant to protect. Of all the reinsurance examined over four years, the authors estimated that fewer than one contract in twenty served any legitimate purpose at all.²

A medical scheme is a not-for-profit entity; its surplus belongs to its members. Yet schemes were entering into reinsurance contracts that moved that surplus out — not to spread risk, but to deliver it to a related party. The pattern was specific and, once seen, unmistakable. The heaviest users were not the small schemes that might genuinely need protection but the largest ones, with tens of thousands of members, comfortably able to absorb their own volatility. The contracts were not modest excess-of-loss treaties trimming the tail; they were large proportional quota shares — eighty, ninety, a hundred per cent — that swept the bulk of the contribution income out of the scheme in a single motion. And the reinsurer, with striking regularity, belonged to the same corporate group as the scheme’s own administrator.

The money left by two doors. The obvious one was the premium itself, on which the schemes booked systematic losses. The subtler and more instructive one was the fee: a profit-share buried in the treaty let an administrator advertise a low, competitive headline charge to the scheme’s trustees and then top it up, out of sight, through the reinsurance leg. The headline price looked reasonable; the real economics sat in the channel the related party controlled. There was no arm’s length anywhere in it, and that was the whole condition of the thing — the same interests sat on both sides of the contract and set its price between themselves, watched over by trustees too weak, too conflicted, or too incurious to object.

Read those paragraphs again with the names changed, and they stop being a story about a small health market and become a description of the largest growth area in life insurance today. The structure now spreading across the United States — and reaching into the British bulk-annuity market through funded reinsurance — has every feature the South African regulator condemned, reproduced at a thousand times the scale. The cedants ceding hardest are the largest balance sheets, not the ones that need protection. The treaties are coinsurance of entire blocks, not tail covers; you cannot move eight hundred billion dollars of reserves with an excess-of-loss layer.³ The reinsurer sits offshore, in Bermuda or the Cayman Islands, and belongs to the same private-capital group as the insurer. And the value leaves, once again, through the fee. The forty per cent with which this essay began is the South African profit-share, grown up and moved offshore.

Even the defences have survived intact. Told its arrangements were extracting members’ money, the South African industry replied that reinsurance let schemes raise capital efficiently and that ceding to a related party aligned everyone’s incentives. Those are, almost to the syllable, the two arguments now advanced for affiliated offshore reinsurance. The actuaries who wrote the 2000 report had already answered them: a large, adequately reserved entity does not need the capital, and genuine alignment can be written into a management contract without routing anything through a reinsurance vehicle at all. The wrapper does not solve the problem it claims to solve. It solves a different problem, for a different party.

*  *  *

There is a domestic version of this, conducted with a different instrument, and the British reader should recognise it. In the bulk purchase annuity market the lever is not an affiliated management fee but the matching adjustment — the regulatory permission to discount long-dated liabilities at a rate above the risk-free curve, and to recognise the resulting gap as value on day one. Funded reinsurance, increasingly to the same offshore jurisdictions, extends the trick. The mechanics differ; the grammar is identical. In each case a regulated entity holding obligations to ordinary people is restructured so that a defined, present, capturable benefit accrues to the owner, while the obligation it was created to honour recedes into a longer and less visible future.

Nor is that defence a museum piece. It is being made now, by the profession’s own funded reinsurance working party, in answer to a Prudential Regulation Authority that has begun — through a 2024 supervisory statement and a pointed 2025 speech — to ask whether the capital these structures save is matched by the risk they actually carry.⁴ The reply runs along three lines: that funded reinsurance is risk transfer and not regulatory arbitrage; that the proportionate answer is better risk management rather than more capital; and that to demand more capital would stifle innovation. Each is defensible as a sentence; none is established as a finding. The load-bearing claim — that the motive is risk transfer rather than the exploitation of a regulatory difference — is exactly the thing a critical paper would have measured, by setting the capital saved against the risk genuinely shed and showing which dominates. It is instead asserted and moved past. And where evidence is offered, it is a poll of the conference hall: the practitioners who write and sell these structures, canvassed on whether the structures should be constrained, their sentiment then reported as though it had settled the matter. The tell is in the remedy. Having half-conceded that the assets largely leave the country, the working party proposes not to constrain the arrangement but to loosen the matching adjustment — to widen the very lever just described, so that more of the day-one value might be captured onshore instead of off. The problem made by one permission is to be solved by enlarging another. None of this is bad faith. It is what results when the people best placed to scrutinise a structure are the people whose work it sustains: the arm’s-length voice missing this time not from the deal but from the deliberation about the deal.

The asymmetry is the whole story, and it is an old one in new clothes. The fee is paid now, in cash, to a party with the discretion to set it. The risk is borne later, in instalments, by parties who did not negotiate it: the policyholders, whose annuities depend on a reinsurer two jurisdictions away, and, behind them, the public backstops that exist precisely because we have decided as a society that pensioners should not bear the full consequence of a structure they never saw. Privatise the fee, socialise the tail. It is the recurring template of the captive market, and it survives because each individual step is defensible, technical, and boring, while only the sum is alarming.

The honest objection is that the cases are not identical, and they are not. A medical scheme is a mutual; extracting profit from it is improper on its face, because the surplus was never the administrator’s to take. A life insurer, particularly one owned by private capital, is a for-profit company, and returning value to its owners is not in itself a scandal. And part of what the affiliated fee buys is real: origination in illiquid and complex credit that the insurer could not cheaply replicate, work that is not a simple rent on assets under management. The offshore structures have genuine substance — genuine managers, genuine private-credit origination, genuine if aggressive asset-liability management — where the South African abuses sometimes had little more than a backdated contract and an agent’s commission.

But notice which way those differences cut. The for-profit form does not make the modern arrangement safer; it makes it harder to stop, because the easy argument — you cannot take a mutual’s surplus — is no longer available. The economic substance does not make it more benign; it makes it more persuasive, and therefore more durable. And the scale does not make it more contained; it makes it systemic. South Africa was arguing over a few billion rand and a loss to members of perhaps four hundred million. The offshore life market is restructuring close to a trillion dollars of reserves — roughly twice the sector’s own capital³ — much of it standing behind annuities and pension transfers, with the public backstops of two continents waiting at the far end of the tail. The precedent is not smaller than the present case. The present case is the precedent with the safety catches filed off.

What makes the South African episode genuinely useful, rather than merely ironic, is that it did not stop at diagnosis. The regulator acted, and bluntly. Reinsurance contracts became illegal unless the Registrar had explicitly approved them, and approval required the scheme to show that the treaty actually spread risk, that no conflict of interest tainted it, and that it served the members’ interest against some identifiable and unusual exposure. It was a heavy instrument. It worked. The conduit closed. The diagnosis and the data cross the decades intact; the cure does not, quite. South Africa could outlaw the structure because almost none of it was doing anything useful, whereas the offshore version is wrapped around genuine credit work that a blunt prohibition would destroy along with the rent. That is not a reason to look away — it is the reason the offshore problem is the harder one, since the camouflage is real and load-bearing. What crosses the decades undiminished is the warning, and the industry is proceeding as though the report had never been written.

And there is a detail in that report that should unsettle the profession most. Its real subject, in the end, was not the schemes or the insurers but the actuaries, who were present in almost every transaction in which reinsurance had been abused — as trustees, as directors, as the consultants who designed the contracts and signed them off. A single scheme in 1999 was already running a cell captive, controlling its own reinsurer with onward cover behind it. That oddity is now the industry’s dominant architecture; we call it a sidecar. The same profession that built the conduit then is building it again, at scale, offshore, and calling it by a newer name.

None of this is settled by the questions the regulators mostly ask — whether the offshore reinsurer is solvent, whether its assumptions are disclosed. Both are fair; neither is the central one. The central question is about incentive: when the same parent owns the cedant, the reinsurer and the manager, whose thirty-year interest is the fee actually serving? The owner is paid on assets gathered and fees extracted, both realised long before the liabilities they stand against come due, while the policyholder cannot afford the indifference to the back end of the book that the owner can comfortably hold. That indifference is not malice. It is simply what the incentive describes, and incentives describe behaviour more reliably than mission statements do.

So the question worth putting to the whole arrangement is not whether it is solvent but whether it is insurance at all — or an asset-gathering business that has found, in the regulated insurer, an unusually patient and unusually captive client. South Africa put that question once, answered it, and closed the conduit. We have the report, the numbers, and the regulation that worked. We appear, all the same, to have decided to learn this one the expensive way.

C.J. Marsden

*  *  *

Notes

1. David Walker, “US affiliates’ outsourcing fees in ‘offshore life re’ grew 40% since 2015, Moody’s finds,” Insurance Asset Risk, 8 June 2026 (reporting Moody’s Ratings research); and David Walker, “Comment — Putting the $16bn outsourcing payday for US lifers’ affiliated managers in context,” Insurance Asset Risk, 12 June 2026, which draws on Insurance Risk Data’s review of Bermudian investment-management agreements between 117 named re/insurers and 84 named asset managers. The $16bn aggregate, the leading-ten figures ($11bn in 2025, up from $5bn in 2015), the fee-floor provision and the board-overlap observation are taken from these two pieces.

2. H.D. McLeod, P.G. Slattery and A.M. van den Heever, “The Use and Abuse of Reinsurance in Medical Schemes,” South African Actuarial Journal 1 (2001): 95–117. The figures attributed to this paper include the roughly fifty-five-fold growth in open-scheme reinsurance premium over 1996–1999; the concentration of 95.7% of that premium in schemes with more than 30,000 beneficiaries; the approximately R3.2bn placed through contracts with parties corporately related to the scheme or its administrator; the estimated R400m total loss to members; and the finding that fewer than 5% of agreements served a legitimate purpose. The account of the regulatory remedy — reinsurance contracts rendered illegal unless approved by the Registrar, conditional on genuine risk-spreading, the absence of conflict of interest, and a demonstrable member interest — is drawn from the same paper and from the regulations made under the Medical Schemes Act, Act No. 131 of 1998.

3. Moody’s Ratings, Offshore reinsurance goes mainstream, raising counterparty risk (2024–25), which placed life reserves ceded offshore at approximately $1 trillion as at end-2023 — around twice the sector’s total capital and surplus; and a subsequent Moody’s report (June 2025) estimating that close to $800bn of reserves had been moved to offshore affiliates since 2018.

4. The defence summarised here is that advanced by the Institute and Faculty of Actuaries’ Funded Reinsurance Working Party in its 2025 commentary responding to the Prudential Regulation Authority — in particular to Supervisory Statement SS5/24 (Bank of England, July 2024) and to remarks by Vicky White, the PRA’s Director of Prudential Policy, at the Bank of America Financials CEO Conference in September 2025. Confirm exact title and publication details before publication.