FINANCE, POLICY & MARKETSPublished by Paul Ivinskas
fc.The Financial CurrentDAILY INTELLIGENCEWhat matters across finance
Deep-dive library

Pension funding: discount rates, asset returns and the promises behind funded status

6 min read · estimatedAI-generated analysis · Methodology
Current version · 1 version · Publication details

First published . This version published .

Initial full research explaining the mechanism, worked examples, competing interpretations and material limitations.

Related research, policy & entities ↓

At a glance

Excerpts from this version
What it covers
A pension funding ratio compares assets with a measured promise. Discount rates, valuation rules, benefit design and cash-flow timing explain why the ratio can change even when no contribution arrives.
0% through article

Tap a dotted-underlined term for a definition; terms are highlighted once per section. Use Aa in the navigation for reading preferences.

In this article

The promise comes before the ratio

A defined-benefit pension promises benefits under a formula rather than merely promising an employer contribution to an individual investment account. Valuation estimates the timing and amount of those benefits using assumptions about factors such as service, pay, retirement and survival. Funded status then compares assets with the resulting liability. The IRS distinguishes these promised-benefit plans from defined-contribution arrangements and explains that required defined-benefit funding involves actuarial calculations. [1]

The question “Is the plan fully funded?” is incomplete until the measurement basis and date are specified. An accounting obligation, a statutory funding target, a PBGC premium funding target and a price to transfer benefits to an insurer answer different questions. A 100% result under one basis does not establish 100% under every other basis, nor does it mean future outcomes are certain.

Discounting can change the ratio without changing the benefit

In a deliberately simplified example, a plan owes one $100 million payment exactly ten years from today. There is no mortality uncertainty, inflation adjustment or new accrual. At a 4% annual discount rate, its present value is $100 million divided by 1.04 to the tenth power, or about $67.56 million. At 5%, present value falls to $61.39 million. The promised payment remains $100 million in both cases.

With $60 million of assets held constant, the funded ratio rises from 88.8% to 97.7%, and the measured deficit falls from $7.56 million to $1.39 million. No money has entered the plan. The lower reported liability reflects a changed valuation rate, not cancellation of benefits. Conversely, a lower rate makes the same future payment more expensive in present-value terms.

Real plans have thousands of expected payments across many maturities. A single displayed rate can summarize a and cash-flow matching exercise. BNY’s 2025 pension note, for example, describes reviewing high-quality corporate-bond curves and cash-flow matching models when setting its U.S. pension discount rate. This is a corporate financial-reporting example, not the statutory funding formula for all plans. [2]

Corporate accounting and contribution requirements use different rules

For U.S. single-employer plans subject to the standard minimum-funding framework, IRS guidance describes segment rates, averaging periods and statutory corridors used to discount benefits; alternatives and special rules exist. These can make the funding rate behave differently from a spot market rate used in corporate accounting. Publication 560 separately explains that the minimum required contribution is a technical calculation, with installment requirements where applicable. [1][3]

PBGC’s variable-rate premium uses unfunded vested benefits and its own premium-funding rules. Its fact sheet describes spot segment rates by default, with an alternative available under the regulations. Consequently, a contribution that changes one reported shortfall does not mechanically eliminate every other charge or measured deficit. The premium target, funding target and accounting obligation must not be relabeled as interchangeable. [4]

Consider a hypothetical $90 million asset pool. If one permitted valuation basis produces a $100 million obligation, the ratio is 90%; if another produces $112.5 million, it is 80%. Both ratios can be arithmetically correct for their stated bases. Selecting the more flattering one without explaining the basis would obscure the underlying promise. The example illustrates comparability, not any actual plan’s regulatory treatment.

Public pensions follow a separate institutional framework

Governmental plans are not subject to Title I of ERISA; state and local law can supply parallel obligations. It is therefore incorrect to apply the corporate ERISA funding framework wholesale to a state retirement system. Public financial reporting and contribution policy also serve distinct purposes. [5]

Salt Lake City Community Reinvestment Agency’s June 30, 2025 financial statements provide a concrete public-plan example. The pension note used a 6.85% discount rate for its Utah Retirement Systems interests because projected plan assets were expected to cover the projected benefits under the stated contribution assumptions. The note also showed how liabilities change at rates one percentage point above and below that rate. The assumption is conditional; it is not a guaranteed investment return. [6]

This example also demonstrates why a public-plan ratio cannot be compared mechanically with a corporate ratio derived from a corporate-bond curve. Different rates, measurement dates and liability definitions can explain part of the gap. An expected-return-based public accounting measurement does not make benefit risk disappear, while a market-rate measurement does not by itself prescribe the next contribution.

Asset returns and liability sensitivity can move in opposite directions

The earlier discount-rate example held assets constant to isolate one mechanism. In practice, bond assets generally change value when market yields change. A plan holding long-duration assets that closely match the benefit cash flows can see assets and liabilities move together. A plan holding shorter-duration assets against long-dated benefits may experience a much larger change in its deficit. , cash flows and valuation conventions prevent a perfect universal match.

For another hypothetical case, start with $90 million of assets and a $100 million liability. If an interest-rate shock reduces assets by $5 million and the measured liability by $10 million, the ratio improves from 90% to 94.4% even though assets lose value. If assets instead decline by $15 million with the same liability change, the ratio drops to 83.3%. “Bonds lost money” alone cannot explain the funded-status outcome.

Growth assets introduce a different uncertainty. An assumed long-run return is not spendable cash today, and a poor return just before large benefit payments can force asset sales. Two plans with the same reported ratio can therefore have different exposure to near-term cash needs, market losses and future sponsor contributions.

Closed, frozen and terminated are not synonyms

BNY’s pension note offers a clear distinction: its U.S. plans were closed to new participants at the end of 2010, while benefit accruals based on service or pay were frozen in June 2015. Closing a plan did not itself stop existing employees earning benefits during that intervening period. Freezing accruals did not eliminate already-earned obligations. [2]

A mature frozen plan can gradually become a pool of assets paying a shrinking but long-lived set of liabilities. An open plan continues adding service and potentially new members. New participants can change demographics and cash flows, but they do not erase an existing shortfall. Likewise, termination or an annuity transfer involves a different legal and economic process from simply closing enrollment.

A useful interpretation remains conditional

The constructive interpretation of an improving ratio is that contributions, returns or risk reduction have strengthened the plan. The cautious interpretation is that the improvement depends heavily on a more favorable discount rate or asset valuation rather than additional resources. A reconciliation of beginning and ending assets and liabilities can distinguish these channels.

Later contributions, actual benefit payments, return experience, assumption revisions and consistent-date sensitivity disclosures can clarify whether progress persists. None of the simplified ratios here determines a sponsor’s legally required payment or a participant’s guaranteed benefit. They explain why pension funding is a comparison of measured resources and promises, with the measurement rules kept visible.

Sources

  1. IRS; Publication 560, Retirement Plans for Small Business; 2025 editionOfficial sourceBack to text: ↑1↑2
  2. BNY; 2025 Form 10-K pension note; December 31, 2025 measurementFiling / reportBack to text: ↑1↑2
  3. IRS; Pension plan funding segment rates; current methodology checked October 4, 2026Official sourceBack to text: ↑
  4. PBGC; Pension Insurance Premiums Fact Sheet; updated May 5, 2025Official sourceBack to text: ↑
  5. U.S. Department of Labor; ERISA Fiduciary Advisor, governmental-plan coverage; checked October 4, 2026Official sourceBack to text: ↑
  6. Salt Lake City Community Reinvestment Agency; financial statements, pension note; year ended June 30, 2025Official source · PDFBack to text: ↑

Flag an error or suggest a correction →Public corrections log →