How a cash balance trust is usually invested

By June the trust has been invested for half a year, or it is about to be invested because a new plan finally has an account open. Owners who are comfortable buying equities in a taxable account often give the pension the same instructions. That impulse treats the cash balance trust as a larger 401(k). It is not. The hypothetical account grows by the interest credit in the document whether the portfolio cooperates or not. The trust is the pile of real assets that has to support the promise. When the pile is invested as if the owner personally kept the upside and could ignore the downside, the formula and the portfolio are in a fight. The formula usually wins, and the owner writes a check.

Sterling Pension Group LLC in West Hartford administers plans. We are not an actuarial firm, not a registered investment adviser, and not the trustee. Independent Enrolled Actuaries certify valuations. They do not pick your funds unless you have separately hired someone who does that work and who happens to be an actuary. This briefing is not tax, legal, investment, or actuarial advice. It explains the mismatch so the investment professional is not briefed with the wrong job.

Two rates that are not the same rate

The interest crediting rate is a plan term. Each year, or each period the document names, the hypothetical account is increased by that rate. Common designs use a fixed rate that stays inside the market-rate rules, or a rate tied to Treasury yields, sometimes with a floor. The credit is not a performance fee and it is not a guess about the S&P 500. Once the document says 4 percent, a 4 percent year is what the account receives even if the bond portfolio earned 1 percent and even if it earned 11 percent.

The investment return is what the trustee's portfolio actually did, after expenses. It lands in the trust. It does not automatically land in the hypothetical account. If the portfolio earns less than the credit, the plan is short relative to the benefit that just grew, and the employer's future contributions rise to fill the gap. If the portfolio earns more, the extra stays in the plan as surplus. Surplus can reduce later required contributions, which is genuinely useful. It is not a distribution to the owner. It is not wages. You cannot wire it back to the operating account because the year went well.

That asymmetry is the whole investment policy. Downside is a cash call against the practice. Upside is trapped inside a qualified trust. A taxable brokerage account gives you both sides. A 401(k) gives the participant both sides, which is why aggressive investing can be a rational 401(k) choice for an owner who will not touch the money for twenty years. A cash balance plan with a fixed credit gives the participant a smooth benefit and gives the sponsor the volatility. Aggressive investing fights the interest credit because the credit will not fall in a down year to match the market, and the owner does not pocket the up year outside the plan.

Fees sharpen the point. If the credit is 4 percent and the all-in investment cost is 1 percent, the portfolio has to earn about 5 percent before fees to keep pace with a promise that does not charge itself that fee. A portfolio built to "beat the credit by a little" that ignores expenses is a portfolio built to fall behind. Ask the adviser for the net return the policy is aiming at, and put the crediting rate next to it. If the target is several points above the credit, ask who pays when the several points arrive in the wrong direction.

How these trusts are usually invested, and what the IRS does not require

There is no IRS asset-allocation table for cash balance plans. Anyone who tells you the Service requires a bond ladder is selling a preference as a rule. The usual practice, which is not the same thing as a statute, is to invest so the assets have a reasonable chance of earning something near the crediting rate with a limited risk of a large hole. That often means high-quality fixed income, stable value where it is available and appropriate, Treasury or agency exposure, and sometimes a modest equity sleeve. The duration of the bonds is a real decision. A long-duration bond fund can fall hard when yields rise, which is exactly the year a sponsor discovers that "bonds" was not a synonym for "stable."

The purpose of that posture is to keep the contribution predictable. Predictable is the point of pairing a large deduction with a practice that has payroll to meet. A portfolio that is eighty percent equities can look brilliant for three years and then demand a contribution the practice cannot make in the fourth, in order to fund a benefit that never dropped. The enrolled actuary will measure that hole under statutory rates. The measurement will not be softened because the investment committee meant well. Publication 560 describes the sponsor's funding responsibility in the IRS's plain language. It does not list ticker symbols.

Some documents credit the actual rate of return, within the market-rate rules that keep the design a valid cash balance plan. In that design the hypothetical account moves with the portfolio, and the mismatch shrinks. The benefit itself becomes variable. Staff who have accounts feel the same down year the owner feels. The funding volatility changes shape rather than disappearing, because the statutory funding measurement and the document's credit are still not identical. Choosing actual-rate crediting is a document decision, made before the year, with the actuary, and then left alone. It is not a trading instruction the adviser can adopt in June because equities have been strong. If you want that design, say so before the formula is locked. Do not ask the trustee to improvise it.

The cash balance account is hypothetical precisely so this distinction can be explained. The fees and roles page is where the adviser, the trustee, the administrator, and the actuary are supposed to be named as different people. If one firm is doing two of those jobs, the engagement letter should say so. Silence is how the owner assumes the administrator is watching the portfolio and the adviser is watching the minimum funding. Neither assumption is safe.

Surplus, shortfall, and the check you cannot casually reverse

A shortfall becomes a minimum required contribution. For a calendar-year plan the 2025 minimum is generally due September 15, 2026, and 2026's investment result will show up in a later valuation. Quarterly installments can apply when there is a funding shortfall. The investing decision in June 2026 does not rewrite the 2025 contribution that is already determined, but it changes the path of the next one. Owners sometimes take a loss in the trust and then ask whether they can "pause" the credit. The credit is a plan term. Reducing it is an amendment, with anti-cutback limits on benefits already accrued and with notice rules if future accruals drop. It is not a portfolio rebalance.

A surplus is the more seductive problem. The trust is ahead of the hypothetical accounts. The owner feels the money is "really mine" because the practice deducted the contributions that created it. The deduction was real. The reversion is a different transaction. Taking surplus back to the employer when the plan ends can trigger an excise tax under section 4980, 20 percent in certain cases where the statute's replacement-plan or benefit-increase conditions are met, and 50 percent otherwise. Surplus can also be used to fund benefits, including benefits for employees you did not intend to enrich. Neither outcome resembles selling a stock in a taxable account and wiring the cash home. Do not invest for a surplus you have no lawful, priced way to receive.

While the plan is ongoing, surplus can lower future deposits. That is the legitimate benefit of earning more than the credit, and for a sponsor who intends to keep the plan open it is a reason to allow some modest room above the credit rather than to match it to the basis point. It is not a reason to run an equity-heavy policy. The room you wanted becomes a hole if the extra risk shows up as a loss. The illustrations you were shown assumed an interest credit, not an investment heroics scenario. If an illustration showed a lower contribution because someone assumed the trust would earn 8 percent forever, label that page illustrative and ask what happens in a zero year. There is no flat IRS cash-balance contribution cap that protects you from that zero. The $290,000 figure is a benefit limit, explained on the IRS defined benefit limits page. It is not a cushion against investment loss.

The 2026 defined contribution ceilings are a different plan. A $24,500 deferral, an $8,000 catch-up, an $11,250 catch-up at ages 60 through 63, and $72,000 of annual additions were set by IR-2025-111 and Notice 2025-67 on November 13, 2025. Those limits govern the 401(k) side. They do not give the cash balance trust permission to be invested like a participant-directed 401(k). The compensation cap of $360,000 likewise limits pay the formula can see. It does not limit the investment loss the sponsor must fund. The IRS cost-of-living page is the checklist for those ceilings. It is silent on ticker symbols, which is the correct silence.

Prohibited transactions and the title on the account

The trust is a separate pool. It has its own accounts. It should not be titled in your personal name, and it should not be the operating account with a spreadsheet beside it. You cannot borrow from it to bridge payroll. You cannot use it as collateral for the practice line of credit. You cannot buy a piece of equipment the practice will use. You cannot sell a building you own to the plan so the practice can "pay rent to itself." Those are prohibited-transaction facts, and the tax on them is not a mild administrative fee. Collectibles, a vacation property, and a loan to a partner are in the same family of mistakes. If an adviser proposes something that would be unlawful in an IRA, assume it is at least as sensitive here until counsel says otherwise.

A fidelity bond is required for people who handle plan funds. The Form 5500 asks about it. The bond is not the same thing as fiduciary liability insurance, and neither one is the investment policy. Confirm the bond exists and that the amount still fits the assets. Confirm who the trustee is. A directed trustee follows investment instructions. Someone has to be the person who gives them, in writing. "We all talk" is not an investment policy statement.

PBGC coverage is independent of how you invest. Coverage is plan-specific. Many small professional-service plans are exempt. Not all. Exemption is not a reward for a conservative portfolio, and coverage is not a penalty for owning stocks. The PBGC's coverage page is the reference. Premiums, if they apply, are a cost the investment return also has to overcome. They are not an asset class.

What to do in the next two weeks

Write down the interest crediting rate in the document and the year-to-date return of the trust, after fees. If you do not know one of those numbers, you do not yet have an investment policy. You have an account.

Ask the adviser, in writing, what loss in a single year the practice could fund without distress, and whether the current mix can produce that loss. Ask the actuary, through the administrator, how a 10 percent asset decline would move next year's illustrative contribution. Label the answer illustrative. It is a stress test, not a prediction, and not a new IRS limit.

Confirm the account title, the trustee, the bond, and that no practice payable is being run through the trust. Then bring the mismatch question to us if the portfolio and the credit are far apart and nobody has said so on purpose. The useful note is the crediting rate, the current allocation, and whether you intended the sponsor to bear equity risk. Many owners, once the question is put that way, did not.