NAIC's $1.2 trillion private credit letter reaches Bermuda's $1.52 trillion reinsurance book

Two figures the NAIC has now placed on the same page should be occupying Bermuda's long-term reinsurers this week: the $1.2 trillion private credit book on US insurer balance sheets that the association sized in its letter to Senator Warren, and the $1.52 trillion in long-term reinsurance capital managed out of Bermuda, with the United States supplying 82 per cent of ceded business and Japan 11 per cent. On their face these are separate conversations, one about assets and one about domicile; the letter is what joins them.

The joining thread is the review process the letter describes. The ratings and valuation reviews inside it decide the capital cost of holding private credit, and the same assets held through a Bermuda reinsurance balance sheet rather than on a US life insurer's own books are subject to the same reviews. A valuation decision taken in Washington does not stop at the water's edge; it changes the number a ceded book has to carry.

The transmission is short and does not require the NAIC to reach outside its remit. The association does not regulate Bermuda; it regulates the assets that Bermuda's long-term reinsurers hold, and most of the ceded business in that market originates in the United States. Tighten a valuation assumption, or change how a private credit holding is looked through, and the balance sheet that absorbs the change first may be the one sitting offshore.

The letter does two things, and the second is easy to lose under the first. It puts a size on the private credit book—$1.2 trillion—and it points at a charge. Both halves matter to a reinsurer, because a perimeter only constrains once someone can measure it, and the asset-side treatment of private credit is what sets the capital cost of a ceded book.

Non-reciprocal jurisdiction is a category rather than a rate. The two measures in the letter work on the same balance sheet from different ends: one prices the assets, the other prices where those assets sit. A private credit review on US life company books and a reinsurance charge on offshore domiciles therefore belong in the same conversation, even though one is an accounting question and the other is a geography question.

The asset question and the domicile question are usually argued in separate rooms. A private credit book of that size is a particular kind of exposure, held by insurers that need a capital figure they can defend to a state regulator and under a ratings review, and the valuation and ratings work the letter describes is the machinery that produces that figure. It is the figure, more than the perimeter, that a ceded book carries.

None of this is new in kind. Offshore reinsurance has been migrating for years, and we have argued that the re-sort toward Bermuda is already underway two years before the NAIC's charge lands. What is new is the deadline: the non-reciprocal jurisdiction charge is due at the end of 2027, and the house reading of it is that it re-sorts offshore reinsurance rather than shrinking it.

The split inside that $1.52 trillion deserves more attention than it has had. The United States supplies the dominant share of ceded business and Japan the minority, which makes Bermuda's long-term reinsurers predominantly American cedents with a Japanese minority rather than a genuinely diversified international book. A minority cession is unlikely to be underwritten on its own terms; it is priced off the balance sheet it joins, so the American cessions set the price for the Japanese minority, and Tokyo can renegotiate what the ratio produces, not the ratio.

Set against that market total, the Japanese piece is on the order of $167 billion and the American share well over $1 trillion—figures for scale rather than for a ledger, since the split is of ceded business while the market total is of managed capital. Either way the Japanese pool is far too large to be left unpriced and far too small to set terms on its own.

The two rulebooks meet on that split. The charge takes its name from reciprocity, and a reciprocity test compares regimes: whether one jurisdiction's solvency and reinsurance rules resemble another's. Tokyo holds the minority share, while Washington's asset rules and Washington's standard govern the majority book. A Bermuda reinsurer's blended cost of capital draws on both, weighted by the split, and the Japanese piece is worth pricing now rather than when the charge is closer.

Record capital is part of why the deadline matters rather than resolving it. Reinsurer capital stands at a record $785 billion, and capacity that plentiful does not retreat when a charge appears; it relocates. The argument the charge invites is that an assessment on non-reciprocal jurisdictions, switched on at the end of 2027, changes where business is ceded more than how much of it is.

For a Bermuda long-term reinsurer, the operative variable between now and then is what the NAIC writes down when it defines reciprocal. If the test is applied jurisdiction by jurisdiction, then the treatment of Japanese cessions has consequences well beyond Tokyo: where blended capital cost lands once the charge is live, and whether the marginal dollar of long-term US liability is better held in Bermuda, somewhere else offshore, or back onshore.

The sequencing argues for treating the end of 2027 as nearer than the calendar suggests. A charge with a fixed effective date starts showing up in multi-year terms before it takes effect, so the re-sort is happening now rather than in 2028. By the time the charge is live, many of the cessions it is designed to price will already have been agreed.

The asymmetry is worth sitting with. Washington matters to a Bermuda reinsurer on both sides of the balance sheet, supplying most of the ceded business and writing the asset-side rules the Warren letter will apply. Tokyo matters for the minority share, but at a moment when the NAIC is deciding which jurisdictions carry a charge, that minority is too large to leave out of the pricing, and the Japanese rulebook has to be read alongside the American one.

The practical question, then, is how to price the mix. A Bermuda reinsurer with a dominant US book and a Japanese minority cannot price the two halves separately when it comes to its own cost of capital; the mix is the base. If Tokyo's rulebook is treated as close enough to Washington's, the minority stays cheap and the re-sort runs its course. If it is treated as far enough away to carry its own charge, then the cheapest home for that piece is likely somewhere that is neither Bermuda nor Japan, and the whole book wears whatever blended cost follows.

The charge is due at the end of 2027. Before it lands, the NAIC will have to write down what it means by reciprocal, and the first place that definition shows up will be the cession mix of a Bermuda balance sheet that currently sits at 82 per cent United States and 11 per cent Japan.