Milliman published two funded-status readings for the same measurement date, July 31, 2026, seventeen days apart. The 100 largest corporate defined benefit plans ended the month 112.1% funded with a $139 billion surplus. The 100 largest public plans ended the same day 88.2% funded with an $816 billion deficit.
Corporate assets returned negative 1.55% in July. Public assets returned negative 0.1%. The system that did worse on investments is the one whose funded ratio rose.
Key Takeaways
- A 23.9 point gap on one measurement date between two indices of 100 large US plans is not an investment result. Both sides lost money in July.
- The corporate index gained $25 billion of funded status while its assets fell $27 billion. The liability fell $52 billion, and that is the entire story of the month.
- A 41 basis point discount rate move did it, from 5.61% to 6.02%. The implied effective duration on the corporate liability is about 10.5 years (computed from the $52 billion decline against a $1.208 trillion opening obligation).
- The public index cannot respond to that move at all. Its median assumed return of 7.0% is an assumption about assets, not a market yield, so a rate rally leaves the liability where it was.
- Milliman's own pessimistic scenario returns the corporate index to 98% by the end of 2027 on a discount rate retracing to 5.17%. A surplus built by rates is unwound by rates.
Two Readings, One Measurement Date
The two indices are constructed the same way, cover the same country and the same month, and disagree by 23.9 points.
| As of July 31, 2026 | Milliman 100 corporate PFI | Milliman 100 public PPFI |
|---|---|---|
| Funded ratio | 112.1% (from 109.5%) | 88.2% (from 88.7%) |
| Funded status | $139bn surplus | $816bn deficit |
| Assets | $1.296tn | $6.095tn |
| Liability | $1.156tn | $6.911tn |
| July investment return | -1.55% | -0.1% (range -2.1% to 1.2%) |
| Discount basis | 6.02% market yield, up 41bps | 7.0% median assumed return |
Sources: Milliman Pension Funding Index, August 10, 2026; Milliman Public Pension Funding Index, August 27, 2026.
Neither index had a good investment month. The corporate side had the worse one by 1.45 points and gained 2.6 points of funded ratio. The public side had the better one and lost half a point.
The $25 Billion Came Off the Liability
The corporate index lost $27 billion of assets in July and shed $52 billion of projected benefit obligation. The net was a $25 billion improvement in funded status, and every dollar of it came from the discount rate moving from 5.61% to 6.02%.
That is the mechanism the two systems do not share. Corporate plans reporting under ASC 715 discount benefit cash flows at yields on high-quality corporate bonds, so the liability is a bond-like quantity that reprices monthly. Public plans reporting under GASB 67 and 68 discount at the long-term expected return on plan assets for the portion of benefits projected to be covered by assets, blending in a municipal bond index rate only past the crossover point. In practice the reported rate is the return assumption, and Milliman's index carries a 7.0% median.
An assumption does not move when the Treasury curve moves. The public index's $6.911 trillion liability was insensitive to the best month for discount rates in the past year, which is why a 6.1% year-to-date asset return through July still leaves it at 88.2%.
The consequence for a corporate sponsor is that the funded ratio is only as durable as the asset side's rate sensitivity. A liability with roughly 10.5 years of effective duration paired with an asset portfolio carrying materially less has a surplus that exists at one point on the curve. The 2.6 point gain in July is the same trade running favourably that would run against the plan on a 41 basis point rally, and a sponsor reading 112.1% as capacity to take investment risk has the causation backwards. The month proved the plan is short duration, not that it is comfortably funded.
For a public sponsor the reverse constraint binds. The funded ratio cannot be improved by a rate move, only by asset returns above 7.0% or by contributions, and the distribution is widening rather than converging: 49 plans now sit above 90% funded, down from 50 in June, while 11 sit below 60%, up from 10.
Milliman's Own Pessimistic Case Takes It Back
The corporate surplus is one rate cycle deep. Milliman publishes the scenario itself: under a pessimistic case of 2.61% annual returns and the discount rate retracing to 5.77% and then 5.17%, the index falls to 108% at the end of 2026 and 98% at the end of 2027. A 41 basis point rise produced the July gain; roughly two of those in reverse erases the surplus entirely.
The base case is not a contradiction of that, it is the same point stated calmly. Holding the discount rate flat at 6.02% with 6.61% returns and $15 billion of annual contributions, the index reaches 112.5% at the end of 2026 and 113.5% at the end of 2027. Almost all of the projected improvement over eighteen months is smaller than the single month of July, because the base case assumes the rate does not move again.
That is what makes the timing question live rather than academic. A plan at 112.1% can settle obligations without contributing cash, and settlement locks the ratio at a level that currently depends on a discount rate holding. A plan that waits is not holding a surplus; it is holding an unhedged position on the long end that happens to be showing a gain.
The public plans have no equivalent decision to make, which is the sharpest thing the pairing reveals. Their 88.2% is a statement about assets and an assumption. The corporate 112.1% is a statement about assets and a market. Two numbers 23.9 points apart, measured the same day, are answering different questions, and only one of them can be changed by the bond market between now and the next reading.
Further Reading on actuary.info
- Corporate Pensions Close Q2 2026 in Surplus: The Endgame Math - the June 30 reading of the same corporate index, one month before this one.
- DB Plans Hit 109% Funded: PBGC Premium and PRT Pressure - what a sponsor does with a surplus once it is large enough to act on.
- Multiemployer Pensions Hit Record 106% Funded at Midyear 2026 - the third US system, on a third set of funding rules.
- Corporate DB De-Risking: Glidepath Triggers in 2026 - the mechanism that converts a funded-ratio print into an asset allocation change.
- PRT Cost Falls Below ABO: The DB Sponsor Inflection - the settlement price a surplus plan is deciding against.
- Milliman April 2026 PBI: PRT Buyout Cost Falls to 101.1% - the buyout index that tracks alongside the funding index.
- Longevity Swaps Fill the De-Risking Gap for Plans Too Large for the Buyout Market - the option for sponsors who cannot settle in one transaction.
- 2027 DB Mortality Tables: Minimum Lump Sum Mechanics Under the SECURE 2.0 Cap - the other assumption set that moves a corporate liability.
Sources
- Milliman, "Pension Funding Index August 2026" (August 10, 2026) - the 112.1% funded ratio, $139bn surplus, $1.296tn assets, $1.156tn PBO, the $27bn asset and $52bn liability moves, the 5.61% to 6.02% discount rate, the -1.55% July return, and the base and pessimistic forecast scenarios.
- Milliman, "Public Pension Funding Index August 2026" (August 27, 2026) - the 88.2% funded ratio, $816bn deficit, $6.095tn assets, $6.911tn liability, the -0.1% July return and plan range, the 6.1% year-to-date return, the 7.0% median assumed return, and the plan counts above 90% and below 60%.
We are seeking feedback on how to improve the site and deliver high-quality content relevant to actuaries. Help us make it better.
Stay ahead with daily actuarial intelligence - news, analysis, and career insights delivered free.
Subscribe to Actuary Brew Browse All Insights