What looks like diversification can, in the wrong hands, become a disguised concentration. That is the unnerving lesson now hanging over private credit’s marriage to the insurance industry. The latest alarm is not that the system has run out of money, but that it has found a cleverer way to move risk around while making everyone feel safer. In markets, as in nature, the most dangerous predator is often the one that blends into the landscape.
A recent FT story argued that private credit’s insurance boom could have hidden costs, calling a “democratised” financial crisis still a crisis. That is the right instinct. The modern financial system loves to rename fragility as access. It does not eliminate danger; it only spreads it across more balance sheets, more legal structures, and more layers of faith in ratings, regulation, and model-based reassurance. The result is less a fortress than a shell: impressive from a distance, brittle when struck.
The newest criticism centers on a paper by Andrew Granato of UT Austin Law and Pranjal Drall of Yale Law, published on SSRN around July 21, 2026. Their argument is blunt: private equity firms such as Apollo, Blackstone, and KKR use life-insurer premium income to fund private credit assets. That is not a crime of arithmetic. It is a design choice. And design choices in finance matter because they determine where losses finally land when assumptions stop behaving.
The paper’s most arresting claim is that taxpayers bear about 86.5% of the final cost of an insurer insolvency through state guaranty funds and insurance-tax deductions. That figure is the authors’ model, not a regulator’s verdict, so it should be treated with care. Still, the point is larger than the estimate. If the structure is built so that gains are privatized and the tail risk is socialized, then the system is not truly dispersing risk. It is hiding the last mile of risk transfer behind a public wall.
This is the old moral hazard problem wearing a modern suit. Once upon a time, banks borrowed short and lent long. Today, the shadow may sit inside insurance companies, where policyholders and state backstops help furnish the funding base. The form is different, but the logic is familiar. When institutions can earn private returns while pushing the consequences outward, the system becomes less like a market and more like a relay race in which the final runner is the taxpayer.
UBS Chairman Colm Kelleher has warned of “looming systemic risk” from weak US insurance regulation and “huge rating agency arbitrage.” He added, “We’re beginning to see huge rating agency arbitrage in the insurance business… In 2007, subprime was all about rating agency arbitrage.” He also said, “If we look at the insurance business, to me, there is a looming systemic risk coming through and it’s because of lack of effective regulation.” His words matter not because they are elegant, but because they point to a recurring weakness: institutions often mistake a rating for an analysis and an analysis for a guarantee.
That weakness is ancient. In game theory, players do not merely respond to incentives; they respond to the structure of payoffs. If a system rewards firms for reaching an acceptable rating while allowing the underlying risk to deepen, then the rating becomes the objective and not the byproduct. That is how arbitragable rules corrode discipline. The rule is satisfied, the danger remains. The paper shield is intact right up until the spear arrives.
There is also a deeper psychological trap. Investors tend to prefer risks they can name. “Insurance assets,” “private credit,” “ratings,” “capital efficiency” — these are soothing labels. They suggest control, professionalism, and a measured distribution of exposure. But the real question is not what the asset is called. It is what happens when it must be sold, written down, or rescued. Finance is full of polished language that functions like camouflage in the forest: it helps the herd feel hidden even while the hunter is already near.
Apollo CEO Marc Rowan rejected Kelleher’s criticism, saying “Colm is just wrong.” He also said 70% of Athene’s assets carry two ratings from S&P, Moody’s, or Fitch, and that ratings are “not where the focus should be.” On one level, that is a standard defense. On another, it reveals the core disagreement. Kelleher sees a system in which ratings may be too easy to game. Rowan sees a system in which the right actors manage risk responsibly and critics overstate the danger.
Both cannot be right in the same way. The disagreement is not about manners but about whether the structure is fragile or merely controversial. If a machine is operating only because its labels are accepted, then the machine is more exposed than its operators admit. If, however, the structure is robust and the critics are extrapolating from a few high-profile cases, then the fear will prove overstated. Markets often force this test by panic, not by debate. But debate is cheaper.
This is where contrarian thinking becomes useful. The herd usually asks whether the industry is growing. The harder question is what must be true for the growth to remain safe. In this case, a great deal must go right. Regulators must stay alert. Ratings must remain meaningful. Insurers must avoid loading up on complexity for the sake of yield. Asset values must remain stable enough to prevent forced selling. And the public backstop must remain distant enough that no one starts pricing it as free. That is a long list of conditions for a business pitched as prudent.
The scale of the move into private debt is what turns a niche question into a systemic one. Bloomberg, citing CreditSights data, reported that US life insurers allocated close to one-third of their $5.6 trillion in assets to private debt last year, up from 22% a decade ago. That is not a rounding error. It is a structural migration. Once an allocation reaches that size, it is no longer a side pocket. It becomes a pillar. And when a pillar shifts, the roof does not ask whether the move was popular.
The recent market moves also sharpen the concern. Apollo Global Management stock is down 30% year-to-date, as of the Eisman report. Blackstone’s BCRED fund saw record $3.8 billion in redemption requests from its $82 billion fund. Blue Owl Capital halted quarterly redemptions and liquidating $1.4 billion in assets. Each case is different, and none should be lazily merged into a single story. But together they show that the industry is not floating above stress. It is already meeting it, and not always elegantly.
That matters because systems often look strongest when they are most levered to trust. A bridge can stand for years if traffic is orderly. Then one load, one vibration, one missed inspection, and the weakness becomes visible all at once. Private credit’s appeal has always been that it seems to offer yield without the daily humiliation of public markets. Yet illiquidity is not the same as safety. It is often just the absence of a quote.
The most uncomfortable feature of the Granato-Drall critique is not that insurers invest in private credit. It is that the cost of failure may not stay private. State guaranty funds and insurance-tax deductions are meant to protect policyholders and stabilize the system. Those are worthy goals. But every safety net changes behavior. If too much risk can be pushed into structures with a public cushion underneath, then the cushion begins to shape the risk that lands on it. That is how protections become invitations.
History offers a familiar pattern. Before the financial crisis, many institutions treated AAA as if it were a law of nature rather than a human judgment. The lesson was not that ratings are useless. It was that ratings are only as good as the incentives and assumptions behind them. The same lesson applies here. When a system pays for complexity with the promise of diversification, then uses ratings and regulation to certify the result, it may be building a cathedral on compressed air.
None of this proves a collapse is coming. That would be lazy prophecy. The better warning is more austere: systems built on layered confidence can look stable long after they have become less resilient. The danger is not a single dramatic lever. It is the accumulation of small assurances, each reasonable on its own, that together create an illusion of robustness. Fragility rarely arrives wearing a mask that says fragility.
The real question now is whether private credit’s insurance channel is a useful capital machine or a rerouting device for risk that would otherwise be harder to hide. If Kelleher is right, the system is tolerating a familiar pre-crisis pattern under a new name. If Rowan is right, the industry is being judged by the worst actors and the loudest suspicions. Either way, the burden of proof should sit with the structure, not the slogan. Finance has spent too many years confusing scale with safety.
