Corporate Pensions Close Q2 2026 in Surplus: The Endgame Math

The Milliman 100 Pension Funding Index closed June 2026 at 109.5% funded, with $1.323 trillion in assets against $1.208 trillion in liabilities, a $115 billion cushion built largely on a 4.81% second-quarter return (Milliman, June 2026).

A funded ratio compares two independently priced instruments. When both move the sponsor's way in the same quarter, a surplus stops being a snapshot and becomes a position with an exit cost.

Key Takeaways

• $115 billion of surplus is 9.5% of the $1.208 trillion liability base, so a liability-side move of that size with assets flat erases it. At a 10-to-12-year duration, each 10 basis points of rate decline adds roughly $12 billion to $14 billion.
• Pension risk transfer payouts fell to $12.6 billion from $23.4 billion the prior fiscal year, in the same year the aggregate cohort surplus more than tripled from $13.4 billion to $48.1 billion.
• 75 of the 100 plans beat their own expected return, earning 8.80% against a 6.61% long-term assumption, which is the premium a fully duration-matched portfolio gives up.
• IRC Section 4980 taxes a reversion at 20% minimum and 50% without a qualified replacement plan, a $23 billion versus $58 billion spread if the aggregate surplus were reverted unmitigated.
• Milliman's scenario cone runs 95% to 129% funded by year-end 2027, a 20-point spread that is the actual planning range behind the single headline ratio.

The Quarter Did Not Move in a Straight Line

Plan assets gained $41 billion between April and June while liabilities rose only $2 billion, carrying the funded ratio from 106.1% at the start of the year to the 109.5% close.

June broke the pattern. The discount rate ticked down one basis point to 5.61%, returns slowed to 0.42% for the month, and the surplus fell by $2 billion even as the ratio held near its quarterly high. "Subpar investment performance combined with rising liabilities caused the PFI funded status surplus to fall during June," said index author Zorast Wadia.

The full-cohort annual study tells the same story from a longer angle. Across the 100 largest corporate DB plans at fiscal year-end 2025, the aggregate funded ratio rose from 101.1% to 103.8% and the dollar surplus more than tripled from $13.4 billion to $48.1 billion, on $1.301 trillion of assets against a $1.253 trillion projected benefit obligation (Milliman 2026 study).

Investment performance did the work. The cohort earned an actual 8.80% against a 6.61% expected long-term assumption, with 75 of 100 plans beating their own target, while a discount rate falling 8 basis points from 5.39% to 5.31% worked against the surplus. Two consecutive years of surplus changes what an actuary solves for: not minimum required contributions and PBGC exposure, the frame in record-high funded status and premium pressure, but a choice among end states.

What the Surplus Is Actually Sensitive To

The $115 billion equals roughly 9.5% of the liability base. Large frozen and closed corporate plans typically carry effective liability durations of 10 to 12 years, so each 10-basis-point decline in the discount rate raises the liability by roughly 1.0% to 1.2%, or $12 billion to $14 billion at the current PBO. On that arithmetic, an 80-to-95-basis-point drop with no offsetting return erases the cushion.

Wadia named the compounded version as the real exposure rather than an isolated bad month, pointing to "sustained, declining interest rates, coupled with a market downturn" (PLANSPONSOR, June 2026). Milliman's pessimistic leg embeds only a 30-basis-point decline paired with a 2.61% return path and still lands at 105% for year-end 2026, because the return assumption rather than the rate move dominates.

Liability-driven investing closes that gap by holding fixed income whose duration mirrors the liability, so a rate move inflating the liability produces an offsetting mark-up in the bond portfolio. The cohort is rotating slowly: equity fell to 24.1% of assets from 24.7% and fixed income rose to 52.7% from 52.2%. That is a half-point move, not a completed glide path, and full immunization would require fixed-income duration matched to the 10-to-12-year liability, materially longer than a core-bond benchmark, which is why these glide paths lean on long corporate credit and STRIPS.

The price of that protection is the return premium itself, the gap between 8.80% actual and 6.61% expected. For a sponsor already committed to termination, giving it up costs almost nothing, since the assets will not be held long enough to compound. For a sponsor choosing hibernation over years, it is a live wager with the rate arithmetic above setting the downside.

Settlement accounting decides when this hits earnings. Under ASC 715 a settlement triggers once the irrevocable transfer of obligations exceeds annual service cost plus interest cost; below that the accumulated gain or loss in AOCI can be amortized, above it the pro-rata share is recognized at once.

That makes partial de-risking an accounting choice as much as a risk one, and the volume shows it. PRT payouts totaled $12.6 billion, down from $23.4 billion, in the year the surplus tripled. Sponsors treated the gains as provisional rather than a cue to lock in buyout pricing.

The Exits That Capture the Surplus Are the Ones That Cost

Path	Mechanism	Surplus treatment
Hibernation	Plan stays frozen; LDI and de-risking triggers manage the surplus in place	Retained, subject to ongoing PBGC premiums and market risk
Thawing	Plan reopened, often redesigned as a cash balance formula	Redeployed into new accruals rather than distributed
Termination	Full annuity buyout or lump-sum settlement closes the plan	Reverted to sponsor (taxed) or transferred to a QRP or benefit increase
Section 420 transfer	Excess above 125% funded moved to fund retiree health or life benefits	Repurposed, one transfer per taxable year

Termination is where the sensitivity meets tax law. Under IRC Section 4980, surplus assets reverting to the sponsor face a 20% excise tax at minimum, rising to 50% without a qualified replacement plan or participant benefit increases. Relief from the higher rate requires transferring at least 25% of the potential reversion into a QRP, or applying at least 20% of the excess to benefit increases inside the terminating plan (Mercer).

On the $115 billion aggregate, the gap between those rates is roughly $23 billion against $58 billion of tax leakage if reversion were pursued unmitigated. That is why full terminations rarely target a large reversion and instead settle close to par, which leaves buyout pricing rather than the excise tax as the binding cost.

Thawing avoids the tax by never letting the assets leave the trust. IBM's return to accruals used a cash-balance design rather than reopening a final-average-pay formula, a structure now covering roughly half of active DB plans against about 15% in 2010 (Milliman). The Section 420 transfer is the narrowest option: available only above 125% funded, capped at one transfer per taxable year, and earmarked for retiree health or life benefits rather than general use.

Milliman's cone runs 105% to 116% by year-end 2026 and 95% to 129% by year-end 2027, and the pessimistic leg puts the ratio back below 100% inside 18 months. A surplus that can invert that fast is one where every exit route carries a toll measured against a cushion that may not still be there when the toll comes due.