Actuarial cost is the present-value estimate of all future benefits a pension plan or insurance arrangement expects to pay, translated into the periodic contributions or reserves needed to fund those obligations over time. For a defined benefit pension, that figure usually splits into two parts: the normal cost that accrues for the current year of employee service, and any payment toward unfunded past obligations. The number matters because it drives both the cash an employer must contribute and the liability it reports on its balance sheet.
What Actuarial Cost Means
At its core, actuarial cost converts a long stream of future payments into a lump sum expressed in today’s dollars. A pension plan promising monthly checks to thousands of retirees over the next 30 to 40 years needs a way to measure that total commitment now, so the sponsor knows how much to set aside each year. Two principles make the translation possible.
The first is the time value of money. A dollar the plan will pay a retiree in 2046 costs less than a dollar today, because money invested now will grow. Actuaries apply a discount rate to pull every future payment back to present value. The second principle is probability. Not every employee will survive to retirement, stay with the company long enough to vest, or live to any particular age after retiring. Mortality tables, turnover rates, and retirement-age assumptions adjust each projected payment by the likelihood it will actually come due.
The resulting figure serves two distinct purposes. The funding cost determines how much cash the employer must contribute each year to satisfy federal minimums under the Employee Retirement Income Security Act.{1Office of the Law Revision Counsel. 29 U.S. Code 1082 – Minimum Funding Standards} The accounting cost determines the pension liability on the corporate balance sheet under generally accepted accounting principles, specifically FASB Accounting Standards Codification Topic 715. The two calculations start from the same obligation but use different assumptions and methods, so they routinely produce different numbers.
The Assumptions Behind the Number
Every actuarial cost figure is only as reliable as the assumptions behind it. Small changes in key inputs can swing the result by millions of dollars for a large plan. The assumptions fall into two groups.
Economic Assumptions
The discount rate is the single most powerful lever. A higher rate shrinks the present value of future payments, producing a lower liability. A lower rate does the opposite. For funding purposes, federal law requires single-employer plans to use three “segment rates” derived from 24-month averages of high-quality corporate bond yield curves, not the plan’s own expected investment returns.{2eCFR. 26 CFR 1.430(h)(2)-1 – Interest Rates Used To Determine Present Value} The three segments cover benefits payable within five years, within 5 to 20 years, and beyond 20 years, and the IRS publishes updated rates monthly.{3Internal Revenue Service. Pension Plan Funding Segment Rates}
For accounting purposes under ASC 715, the discount rate takes a different approach. It reflects the rate at which the pension obligation could effectively be settled, measured by looking at returns on high-quality fixed-income investments such as corporate bonds rated Aa or higher. Salary increase assumptions also matter for plans that base benefits on final or career-average pay, since higher projected raises mean larger future benefit checks.
Demographic Assumptions
Demographic assumptions project the human side: when employees will retire, how many will leave before vesting, and how long retirees will live. Mortality tables carry particular weight. For 2026 funding valuations, the IRS requires plans to use the static mortality tables specified in Notice 2025-40, built from updated base mortality rates and mortality improvement projections.{4Internal Revenue Service. Updated Static Mortality Tables for Defined Benefit Pension Plans for 2026} A separate unisex table blending 50% male and 50% female mortality rates applies when calculating the present value of certain lump-sum distributions.
Turnover assumptions reduce the projected obligation by accounting for workers who will leave before earning full benefits. Retirement-age assumptions affect when payments begin. These inputs typically start from the plan’s own experience studies supplemented by industry-wide data, and they get updated as actual experience diverges from the projections.
How the Cost Gets Spread Across Years
Choosing assumptions tells you the total size of the obligation. Choosing a cost method tells you how to spread that total across each year of an employee’s career. Two methods dominate pension practice, and they produce noticeably different annual cost patterns.
Projected Unit Credit
The projected unit credit method assigns cost based on the specific benefit earned in each year of service. For a plan paying 1.5% of final average salary per year of service, PUC calculates the present value of the slice attributable to this year’s service, using projected rather than current salary. Because each year’s unit of benefit is discounted over a shorter remaining period as the employee ages, the normal cost under PUC starts relatively low for younger workers and grows over time. That rising cost pattern is the method’s defining feature.
GAAP accounting under ASC 715 generally requires a benefit-attribution approach consistent with PUC for salary-related plans, because it matches cost to the period in which benefits are earned. That alignment makes it the standard for balance-sheet reporting.
Entry Age Normal
The entry age normal method takes the total projected benefit obligation and spreads it as a level percentage of pay, or a level dollar amount, from the employee’s hire date through retirement.{5Actuarial Standards Board. Entry Age Normal Actuarial Cost Method} Because cost is leveled from the start, it produces more stable year-to-year contributions than PUC. Plan sponsors often prefer EAN for budgeting and cash-flow planning, and it is commonly used for funding valuations.
Both methods arrive at the same total cost over the life of the plan. The difference is timing. PUC back-loads cost toward the later years of a career; EAN distributes it evenly. When a plan’s workforce is aging, PUC tends to produce higher near-term costs than EAN would for the same obligation.
Funding vs. Accounting: Why Two Numbers
Federal law sets the floor for annual contributions. The minimum required contribution for a single-employer plan equals the target normal cost for the year, plus any shortfall amortization charge, plus any waiver amortization charge.{6Office of the Law Revision Counsel. 26 USC 430 – Minimum Funding Standards for Single-Employer Defined Benefit Pension Plans} If plan assets already equal or exceed the funding target, the minimum drops to the target normal cost reduced by the surplus.
The shortfall amortization charge is where underfunding becomes expensive. When a plan’s funding target exceeds its asset value, the gap creates a shortfall amortization base that must be paid off in level annual installments. The Pension Protection Act of 2006 originally set a seven-year schedule, but the American Rescue Plan Act of 2021 extended it to 15 years for plan years beginning after 2021. The longer window gives sponsors more breathing room on annual contributions but keeps the shortfall on the books longer and accumulates more interest along the way.
The pension liability on the company’s balance sheet is not the same number. GAAP under ASC 715 typically requires a projected benefit obligation measured using a discount rate tied to high-quality corporate bonds rated Aa or above. The funded status reported on the balance sheet is simply the difference between that obligation and the fair value of plan assets. Funding valuations, by contrast, use the three IRS segment rates and may apply different mortality and retirement assumptions. A plan can look reasonably well-funded for regulatory purposes yet show a significant liability on the balance sheet, or vice versa.
When Reality Differs From Assumptions
No set of assumptions perfectly predicts the future, so every year produces a gap between what the actuary expected and what actually happened. These gaps are actuarial gains and losses, and they are a normal part of running a plan.
An actuarial loss increases the obligation. Common causes include retirees living longer than the mortality table predicted, asset returns falling short of the assumed rate, or a decision to lower the discount rate to reflect current market conditions. An actuarial gain works the other way: higher-than-expected investment returns, higher turnover reducing the number of people who earn full benefits, or an increase in the discount rate.
Under GAAP, these gains and losses do not hit the income statement all at once. ASC 715 allows companies to defer recognition into accumulated other comprehensive income and amortize the excess only when the cumulative net gain or loss exceeds 10% of the greater of the projected benefit obligation or the market-related value of plan assets. Amounts inside that corridor can remain unrecognized indefinitely. The smoothing mechanism prevents large swings in reported pension expense from year to year, though it also means the income statement can understate the real economic volatility of the plan.
What Underfunding Costs
Missing the minimum required contribution triggers consequences beyond simply owing the money later. The IRS imposes an excise tax of 10% of the aggregate unpaid minimum required contributions for a single-employer plan.{7Office of the Law Revision Counsel. 26 U.S. Code 4971 – Taxes on Failure To Meet Minimum Funding Standards} If the shortfall still isn’t corrected within the taxable period, a second-tier tax of 100% of the unpaid amount applies. That is not a typo. The penalty for persistent underfunding is a dollar-for-dollar tax on the entire shortfall. Multiemployer plans face a lower initial excise tax of 5% on their accumulated funding deficiency, but the same 100% second-tier penalty applies if the deficiency goes uncorrected.
Every single-employer defined benefit plan also pays premiums to the Pension Benefit Guaranty Corporation, which guarantees benefits if a plan terminates without enough assets. For 2026, single-employer plans owe a flat-rate premium of $111 per participant regardless of funding level, plus a variable-rate premium of $52 per $1,000 of unfunded vested benefits, capped at $751 per participant.{8Pension Benefit Guaranty Corporation. Premium Rates} A fully funded plan pays only the flat-rate premium. A badly underfunded plan with thousands of participants can face variable-rate premiums in the millions, creating a direct financial incentive to improve funded status beyond what the minimum contribution rules alone would require.
Late premium payments accrue interest at the rate the IRS charges for late tax payments, compounded daily. For the first quarter of 2026, that rate is 7%; it drops to 6% for the second quarter.{9Pension Benefit Guaranty Corporation. Late Premium Payment Interest Charges} Additional penalty charges may apply on top of the interest.
Actuarial Cost Outside Pensions
Pension plans get most of the attention, but actuarial cost calculations are equally central to insurance. Life insurers, health insurers, and property-casualty companies all maintain reserves that represent the present value of future claims they expect to pay. The underlying math is the same: discount future cash outflows to present value using appropriate interest rates, then adjust for the probability each payment will actually come due.
For a life insurer, the actuarial reserve on a policy equals the present value of expected future benefit payments minus the present value of future premiums the policyholder will pay. Mortality rates, lapse rates for policyholders who cancel, and assumed investment returns on the insurer’s portfolio all feed into the calculation. Property-casualty insurers face a different challenge. They must estimate reserves for claims that have already occurred but haven’t been fully reported or settled, known as incurred-but-not-reported reserves. Here the actuarial cost depends heavily on historical loss development patterns and assumptions about how long claims take to close.
The regulatory framework differs from pensions. State insurance departments rather than federal agencies set reserve requirements, and the National Association of Insurance Commissioners publishes model standards that most states adopt. The core actuarial question stays the same: how much money needs to be set aside today to cover obligations that won’t come due for months or decades?