Pension Discount Rates Top 6%: Assets Fell $27 Billion and Funded Status Rose Anyway

Corporate pension funded status improved during July 2026 for a reason that had nothing to do with asset performance. The Milliman 100 discount rate rose 41 basis points to 6.02%, its first reading above 6.00% since October 2023, and the projected benefit obligation fell $52 billion while plan assets fell $27 billion (Milliman Pension Funding Index, August 2026).

The index closed the month at a $139 billion surplus and a 112.1% funded ratio, up from $114 billion and 109.5% at the end of June. Zorast Wadia, who authors the index, put the cause plainly: "A large increase in the benchmark corporate bond interest rates resulted in a $52 billion decrease in pension liabilities" (Milliman, August 2026). July is an unusually clean month for isolating what duration does on its own.

Key Takeaways

• 41 basis points of discount rate movement took the Milliman 100 obligation down 4.3%, from $1.208 trillion to $1.156 trillion, an implied effective duration near 10.5 years on the aggregate liability.
• $27 billion of assets left the index in one month on a -1.55% investment return, a 2.0% decline. Funded status still rose $25 billion because the liability fell more than twice as fast in percentage terms.
• 99.7% was July's competitive retiree buyout cost as a share of accounting liability, up from 99.6% in June. The annuity purchase rate rose only 37 basis points against the accounting curve's 41, so the settlement ratio widened.
• $3.8 billion of US pension risk transfer premium closed in the first quarter of 2026, down 47% year over year, while buy-in premium ran the other way, up 443% to $768 million across a falling contract count.
• $111 per participant is the 2026 PBGC flat rate premium, charged regardless of surplus. At 112.1% funded the variable rate premium is already zero, so the $139 billion cushion cannot cut the premium bill further.

The Month's Components: Assets Down 2.0%, Liabilities Down 4.3%

The asset side fell from $1.323 trillion at June 30 to $1.296 trillion at July 31, a 2.0% decline, on a monthly investment return of -1.55% (Milliman, August 2026). The gap between those two percentages is the ordinary drain of a mature book, roughly $6 billion of benefit payments net of contributions in four weeks. The obligation fell from $1.208 trillion to $1.156 trillion, a 4.3% decline.

That asymmetry is the whole month. Every dollar of the $25 billion funded status gain came off the liability, and the index gained ground while losing money. The pattern is not confined to July: year to date, assets are down $10 billion and liabilities are down $74 billion, lifting funded status $64 billion and the funded ratio 6.0 percentage points.

Component	June 30, 2026	July 31, 2026	Change
Discount rate	5.61%	6.02%	+41 bps
Market value of assets	$1.323 trillion	$1.296 trillion	-$27 billion (-2.0%)
Projected benefit obligation	$1.208 trillion	$1.156 trillion	-$52 billion (-4.3%)
Funded status	$114 billion surplus	$139 billion surplus	+$25 billion
Funded ratio	109.5%	112.1%	+2.6 points

Source: Milliman 100 Pension Funding Index, July 2026 and August 2026 editions.

The 6.02% level matters for where it sits in the series rather than for its size. Milliman's last reading north of 6.00% came in October 2023, when the rate reached 6.20% before shedding 120 basis points across November and December to close that year at 5.00%. The twelve months to July 2026 produced a more modest 47 basis point rise, from 5.55%.

The index discounts each plan's projected cash flows on a benchmark corporate bond yield curve, so the entire move is a change in the discounting basis. Nothing in it reflects demographics, accruals, or plan experience.

Implied Duration and What It Did to a Settlement Quote

The release prints the dollar change and the rate change but not the ratio between them, which is the number a sponsor actually needs. A 4.3% obligation decline against 41 basis points implies an effective duration near 10.5 years on the aggregate Milliman 100 liability. At that duration a single basis point is worth roughly $1.2 billion of obligation across the index, or about $10.5 million per $10 billion of projected benefit obligation.

That figure also sets what the month says about hedging. For the surplus to have held flat in dollars, assets would have had to fall $52 billion, or 3.9% of the June balance. They fell 2.0%. A sponsor whose fixed income portfolio genuinely matched the liability's duration captured little of the $25 billion, because a matched book is built so that both sides move together.

One reported ratio conceals that dispersion completely. A closed plan immunised near the full 10.5 years and a sponsor still running a 60/40 allocation both sit inside 112.1%, and the two enter a settlement conversation from opposite positions. The Milliman Pension Buyout Index shows what July did to that conversation: competitive retiree buyout cost rose to 99.7% of accounting liability, from 99.6% in June.

The direction looks wrong until the denominator is read. Milliman reports the annuity purchase interest rate rising 37 basis points in July against the accounting curve's 41 (Milliman Pension Buyout Index, August 2026). The dollar price of a buyout fell; the liability it is measured against fell faster, so the ratio widened. Average pricing moved further, to 103.1% from 102.6%, with competitive bidding estimated to save sponsors around 3.4% as of July 31.

Transfer volume tracks rate levels and insurer capacity rather than funded ratios, which is why the pipeline can thin while the ratio climbs. US pension risk transfer premium totalled about $3.8 billion in the first quarter of 2026, down 47% year over year, with buyouts at $3.02 billion across 85 contracts, down 57% (LIMRA, Q1 2026). Buy-in premium ran the other way, up 443% to $768 million.

PBGC Premiums, the 417(e) Lag, and a Surplus Nobody Can Spend

A funded ratio above 100% stops paying for itself quickly. At 112.1% the plans carry no unfunded vested benefits, so the single employer variable rate premium of $52 per $1,000 of unfunded vested benefits, frozen by Section 349 of the SECURE 2.0 Act, is already zero. What remains is the flat rate premium of $111 per participant for 2026 plan years, up from $106 in 2025 (PBGC, 2026).

That is a head count charge, and another $25 billion of surplus reduces it by nothing. Nor does the surplus reach the sponsor in cash. Absent a plan termination, a Section 420 transfer to fund retiree health benefits, or a reopening that spends the margin on new accruals, it stays an accounting mark, which is why a gain of this size moves contribution policy less than the headline ratio suggests.

Lump sums lag by construction. Section 417(e)(3) minimum present value calculations run on the IRS segment rates for a plan's stated lookback month and stability period, so a calendar year plan using a November lookback fixes its 2027 lump sum factors on November 2026 rates. The third segment rate, which governs the long tail of a lump sum, stood at 6.31% for May 2026 against 6.08% for August 2025 (IRS). July's corporate bond move does not reach most participants' payout elections this year.

The same duration that produced the month can take it back. Milliman's pessimistic path has the discount rate falling to 5.77% by the end of 2026 and 5.17% by the end of 2027, with the funded ratio at 98% (Milliman Pension Funding Index, August 2026). At 10.5 years, 85 basis points of retracement rebuilds roughly $103 billion of obligation, most of the $139 billion cushion. A surplus that cannot be spent can still be erased.

Further Reading on actuary.info

• Corporate Pension Surplus and the Endgame Decision - what sponsors can and cannot do with margin above full funding, path by path.
• Milliman's April 2026 Buyout Index at 101.1% - the same buyout series four months earlier, when the competitive quote still priced above the accounting liability.
• Buy-Ins Overtake Buyouts in the $49B 2025 PRT Market - the structural shift behind the 443% buy-in premium growth.
• Record Funded Status and the PBGC Premium Question - why a flat rate premium per head keeps rising against a shrinking participant count.
• De-Risking Glidepath Triggers in 2026 - how a funded ratio threshold converts a rate move into an allocation change.
• Section 417(e) Segment Rates and Lump Sum Windows - the lookback and stability period mechanics in more detail.