Ask most bookkeepers to close the books on a managing general agent, and they'll reach for the same playbook they'd use for any commission-based business: book the commission when the policy is written, done. That playbook works for maybe 70% of an MGA's revenue. The other 30% — the profit commission that a carrier doesn't even calculate until 18 to 24 months after the policy year ends, and doesn't finalize for up to three years — is where most of the accounting mistakes happen.
If you run or advise a managing general agency, understanding why that delay exists, and how to book the revenue correctly in the meantime, is the difference between financials that reflect reality and financials that quietly overstate your best year and understate your worst one.
What an MGA Actually Does (and Why Its Books Look Different)
A managing general agent operates under delegated authority from an insurance carrier — sometimes called the "capacity provider." Instead of simply referring business the way a retail agent does, an MGA underwrites, prices, and binds policies on the carrier's paper, and often handles claims within a defined program too. In exchange, the carrier grants the MGA a slice of economics that a standard agent never sees.
That economics usually breaks into three layers:
- Base commission — a flat percentage of gross written premium, typically in the 10–25% range depending on the line of business and the MGA's expense load. This is the straightforward part: premium is written, commission is earned, revenue is recognized.
- Ceding commission / expense allowance — compensation intended to cover the MGA's underwriting, policy administration, and claims-handling costs. Also relatively straightforward to recognize as the related services are performed.
- Contingent (profit) commission — additional compensation tied to how profitable the book of business turns out to be, based on the carrier's actual loss ratio, retention, and growth on that program. This is the layer that doesn't behave like ordinary commission income at all.
The first two layers post cleanly against a bordereau — the detailed transaction report an MGA files with its carrier partner, typically split into a premium bordereau (policies written, premium collected, commission due) and a claims bordereau (losses reported, reserves, paid amounts). Reconciling your books against these bordereaux monthly is the single highest-leverage habit an MGA bookkeeper can build, because it's the carrier's own source of truth for what you're owed.
The third layer — profit commission — is where the calendar becomes the problem.
Why Profit Commission Takes 18–24 Months to Even Calculate
Profit commission is calculated by comparing the premium a carrier collected on a program against that same program's incurred losses, expenses, and any ceding commission already paid. The MGA's cut is a pre-negotiated percentage of whatever profit is left over. On paper, that sounds like a simple formula. In practice, it can't be run honestly until claims have had time to develop.
Insurance claims don't announce their final cost the day they're reported. A commercial liability claim opened in month three of a policy year might not settle — or might not even be fully reserved with confidence — until well into the following year. Carriers typically wait until a policy year has "matured" for 18 to 24 months before running a first profit commission calculation, and many treaties explicitly keep recalculating annually until a final, binding settlement at the third anniversary of the policy year, unless both parties agree to extend it further.
That means an MGA writing its 2026 program won't see a first-pass profit commission estimate until sometime in 2027 or 2028 — and won't know the true, locked-in number until 2029. If your books recognize profit commission only when cash hits the bank, you're recognizing 2026 performance in 2029. If your books ignore it entirely until then, your 2026 and 2027 financials understate how the program actually performed, and a strong bind year can look mediocre for two full fiscal years running.
Neither extreme is right. The correct approach is to accrue an estimate — and to be disciplined about how conservative that estimate is.
Booking the Estimate: Constrained Accrual, Not a Guess
Under U.S. GAAP (ASC 606), contingent and profit commissions are a textbook case of variable consideration. The standard requires you to estimate variable consideration and recognize it as revenue only to the extent it's probable a significant reversal won't occur later — commonly applied as a constraint on the estimate, not a ban on recognizing it at all.
For an MGA, that plays out as a two-step process:
- Step 1 — Estimate. Once a policy year has enough claims maturity to support a reasonable loss-ratio projection (often after the first 12 months, refined further at 18–24 months), calculate a preliminary profit commission using the carrier's own formula and your best current loss-development view. Actuarial input matters here — a rough eyeball loss ratio from raw claims data will systematically understate reserves for claims that haven't been reported yet (IBNR, "incurred but not reported").
- Step 2 — Constrain. Apply a haircut to that estimate that reflects genuine development risk — how volatile the line of business is, how much of the policy year's claims tail is still open, and your own program's historical variance between first-pass estimates and final settlements. A property program with fast claim resolution can support a less conservative constraint than a long-tail liability or professional-lines program where claims can take years to fully emerge.
Book the constrained estimate as accrued profit commission revenue, with a corresponding receivable. As each subsequent recalculation comes in — at month 24, then at the final third-anniversary settlement — true up the accrual rather than waiting for a cash reconciliation to surface the difference. Keep a running schedule, by policy year and by program, showing the original estimate, each restatement, and the eventual final number. That schedule does double duty: it's your audit support, and over a few years it becomes your best evidence for calibrating how conservative future constraints should be.
Trust Accounts and Fiduciary Premium: The Other Half of MGA Bookkeeping
Profit commission timing gets the attention because it's unusual, but it isn't the only place MGA books diverge from a normal small business. Because an MGA collects premium on the carrier's behalf, most states require that premium to sit in a segregated fiduciary trust account — never commingled with operating funds, and never treated as agency revenue until the commission actually earned on it is withdrawn per the agency agreement.
Practically, that means every MGA needs at least two clean ledgers that never bleed into each other:
- Premium trust account — carrier's money, held in trust: premium in, carrier remittances and approved commission withdrawals out. Reconciled monthly against the premium bordereau, not just against the bank statement.
- Operating account — the agency's own money: base commissions, ceding/expense allowances, and (once recognized) accrued profit commission.
State insurance departments routinely cite trust account errors — commingling, late remittance, unreconciled balances — as a top audit finding for agencies and MGAs alike. A bookkeeping system that can't produce a clean trust reconciliation on demand isn't just an accounting weakness; it's a regulatory exposure.
A Practical Chart-of-Accounts Starting Point
For an MGA (or an accountant setting up books for one), a workable structure separates each commission type and keeps the profit commission accrual visibly distinct from cash-settled revenue:
Assets:Trust:Premium— fiduciary premium held on behalf of carriersAssets:Receivable:ProfitCommission:<Program>:<PolicyYear>— accrued, constrained profit commission estimate, tracked per program and policy year so restatements are traceable to their sourceIncome:Commission:BaseIncome:Commission:CedingAllowanceIncome:Commission:ProfitCommission:AccruedIncome:Commission:ProfitCommission:TrueUp— separate line for the delta booked when a recalculation lands, so a single year's restatement doesn't get buried inside "normal" commission income
Keeping the accrual and the true-up in separate accounts — rather than one blended "profit commission income" line — makes it obvious to anyone reading the financials how much of this year's number is a fresh estimate versus a correction to a prior year's guess. In a plain-text ledger like Beancount, that separation costs nothing extra to maintain (it's just another account name) but it's the difference between financials a lender or acquirer can actually diligence and financials that require a phone call to explain.
Keep Multi-Year Estimates Auditable, Not Just Accurate
Managing an MGA's books well means holding two things in tension: recognizing profit commission revenue close to when it's actually earned, not years later when cash arrives, while keeping every estimate traceable back to the assumptions and data that produced it. A plain-text, version-controlled ledger is a natural fit for that kind of multi-year accrual tracking — every restatement to a profit commission estimate is a diffable, dated change, not an overwritten spreadsheet cell nobody can explain a year later. Beancount.io gives you that kind of transparent, auditable record for exactly this sort of long-tail accrual accounting, with no vendor lock-in and a full history of every adjustment. Get started for free and see what your books look like when every number has a paper trail.