Insurance by BogdanovaThe Encyclopedia of Insurance and Risk Science

RESEARCH DOSSIER

Insurance Pricing Expense Allocation and Commission Loadings

Abstract. Insurance expenses are the cost of organizing, acquiring, administering, financing, and delivering contingent protection. Their incorporation into premiums is a consequential actuarial choice because the amount, timing, and appropriate allocation basis differ across expense categories, insurance contracts, distribution systems, and jurisdictions. This monograph investigates the economic substance and prospective measurement of insurance operating expenses, the mathematical treatment of percentage commissions, the allocation of fixed and mixed costs, the relation between expense estimates and rate indications, and the effects of remuneration on efficiency and customer value. It integrates the current US actuarial framework, international accounting and prudential distinctions, European insurance-distribution requirements, and established scholarship on intermediaries and transaction costs. It derives transparent gross-premium identities; analyzes policy-based, exposure-based, premium-based, activity-based, and marginal allocations; develops reproducible synthetic illustrations; and tests how commission, fixed-cost, volume, and channel assumptions alter indicated prices. The principal finding is that an expense provision is defensible when its cost definition, observable driver, projection horizon, allocation objective, and contractual incidence agree. Premium-percentage conventions provide useful practical approximations for costs genuinely proportional to premium; they become distorting when applied indiscriminately to policy servicing, underwriting, billing, claims operations, and shared infrastructure. Commission is both an expense and a contractual incentive, so its adequacy must be assessed together with distribution services, risk selection, customer outcomes, and the resulting premium. This work offers original analytical synthesis and constructed examples; it does not claim a new actuarial theory or empirical estimates for any particular insurer.

Keywords: Insurance ratemaking; expense allocation; acquisition expense; commission loading; insurance intermediaries; fixed expenses; variable expenses; cost drivers; activity-based costing; rate indication; expense trend; premium gross-up; expense efficiency; IFRS 17; ASOP 29; insurance distribution; fair value.

Introduction

An insurer prices both the obligation to pay contingent claims and the organizational apparatus that makes the obligation serviceable. Premium collection, underwriting, intermediary relationships, claims administration, customer communication, regulatory compliance, technology, and capital management consume resources before, during, and after the coverage period. These activities differ in their exposure to policy count, claim count, premium volume, policy complexity, geography, distribution channel, and time. An expense ratio calculated against historical premium is an observation about one period's organizational economics; it does not automatically identify the future cost of a contract.

The relevant scientific question therefore has two parts. First, what future resources will the insurer actually consume to acquire and service the expected business? Second, on what economically and institutionally defensible basis should those costs enter the price charged to a class or policy? The ASOP 29 expense standard addresses these questions for prospective US property/casualty risk transfer and risk retention: it requires appropriate categories, intended measures, prospective assumptions, documentation, and consideration of expense estimates on bases other than premium when the costs do not vary directly with it. The ASOP 53 cost-estimation standard situates those components within the broader prospective estimate of transferring or retaining risk.

The investigation proceeds from expense classification to contract remuneration, then to estimation and rate mathematics, and finally to supervisory and distributive consequences. It distinguishes four analytical objects that must remain traceable: economic resources consumed; accounting recognition and reporting; prudential measurement of insurance obligations; and the prospective pricing allowance. A single cash payment can appear differently in these four systems because each answers a different question. The distinction is essential when comparing jurisdictions or explaining apparent discrepancies between actuarial and financial reports.

The monograph is international in interpretation and uses US property/casualty standards as an explicitly identified professional framework rather than universal law. European insurance and prudential rules, UK fair-value requirements, international accounting, and scholarly evidence provide comparisons. Numerical cases are synthetic, their assumptions stated in full, and no rate or commission figure is presented as an industry average. Original contribution lies in the integrated evaluation of cost behavior, distribution incentives, allocation purpose, and the price-equilibrium consequences of charging for expenses, with no proposed new doctrine.

Expense Categories and Cost Bases

Economic purpose and prospective measurement

Expense allocation converts organizational consumption into prospective coverage-level cost estimates. The underlying expense taxonomy must be exhaustive enough for reconciliation, sufficiently distinct to prevent double counting, and operational enough to support a causal or otherwise justified cost driver. Expense measurement begins with actual transactions and activities, but the target is a future contract portfolio, not merely the expenditure of an expired accounting period. A staffing model expected to change next year, a new digital servicing platform, commission contract amendments, shifts in claims volume, and a modified product mix can all make backward-looking ratios misleading.

US actuarial practice identifies commission and brokerage fees, other acquisition expenses, general administrative expenses, loss adjustment expenses, residual-market and statutory-assessment expenses, policyholder dividends, reinsurance costs, and taxes, licenses, and fees as common categories. The current ASOP 29 definitions place insurance acquisition commissions in a separate category from other acquisition activities and require expenses to be projected for the relevant coverage and cost-estimation period. In contrast, investment expenses, federal and foreign income taxes, capital cost, and profit/contingency provisions have distinct placement in that standard's definition of expense provisions. A reconciliation can include these items elsewhere in an overall pricing model, but doing so must preserve the component distinction.

The CAS ratemaking text by Werner and Modlin (2016, ch. 7) uses the principal underwriting expense groupings of commissions and brokerage, other acquisition, taxes/licenses/fees, and general expenses. Its division of other expenses into fixed and premium-variable components is an analytical simplification for the premium equation. A category such as general expense may contain a fixed operating platform, transaction-dependent processing, and a capacity step cost; consequently its label does not establish its economic behavior.

Underwriting, acquisition, claims and financial boundaries

Acquisition expenditures include intermediary compensation, underwriting and policy-issue functions, prospecting, customer acquisition, sales operations, and some technology dedicated to converting business. Their relevant time unit may be an application, quote, bound policy, renewal, or initial contract group. Claims-handling expenses include investigation, validation, negotiation, defense, settlement, claims administration and associated systems. The same claims expense must not appear simultaneously within projected loss and loss adjustment expenses and in a separate general-operating expense allowance. This is a practical application of the LAE boundary in ASOP 29 and the separate prospective cost elements recognized by ASOP 53.

Premium taxes and many statutory charges are often based directly on written or collected premium, although the base, exclusions, rates and timing depend on the jurisdiction. A per-policy fee or flat assessment instead behaves differently. Reinsurance is another distinct boundary: a ceding commission may reimburse the cedant for acquisition or operating expenses and may be conditional on loss experience, but the net retained economics also depend on reinsurance premium, recoveries, reinstatements and profit-sharing. The reinsurance guidance in ASOP 29 expressly calls for a review of these cash flows where reinsurance is included as an expense provision.

At the organizational level, a ledger allocation must reconcile to the expense universe being priced. Write the expected expense pool as E = E_acquisition + E_service + E_claims + E_tax + E_assessments + E_reinsurance, with each term defined so categories do not overlap. Any costs excluded because they belong to the loss provision or a capital/profit account must be identified outside E. The correct partition depends on the insurer's rating model and the applicable accounting and regulatory definitions; the invariant requirement is that total relevant prospective costs be included exactly once.

Expense behavior, units and allocation objectives

The conventional fixed–variable distinction is a relationship to a specified driver, range and horizon. A cost may be fixed with respect to written premium over the next quarter, variable by policy count over a year, and stepwise as staff or computing capacity expands over several years. Let expected expense for portfolio segment i in period t be

Eit=Fit+atNit+btXit+ctMit+dtCit+qtPit+Uit,E_{it}=F_{it}+a_t N_{it}+b_t X_{it}+c_t M_{it}+d_t C_{it}+q_t P_{it}+U_{it},

where F represents allocated capacity and overhead, N policy or transaction count, X exposure units, M measured operating activities such as underwriting hours, C expected claims or claims-handling events, P premium, and U genuinely unclassified or uncertain expense. Every coefficient must have its own consistent unit. For example, a is dollars per policy; b dollars per exposure unit; c dollars per processing minute or hour; d dollars per claim; q a dimensionless premium proportion. This general cost accounting identity is a descriptive representation of established cost-driver practices, not a new actuarial theory. It is useful only when drivers are observable and costs are assigned without duplication.

There are multiple legitimate allocation purposes. Full average cost assigns the entire recurring cost of an organization across an expected portfolio and supports long-run capacity funding. Incremental cost asks which expenses change if a specified contract, channel, or cohort is added; it informs economic decisions but cannot alone sustain a portfolio with common costs. Statutory allocation may prescribe categories or disallow cost components. Financial-statement allocation is designed for recognition and comparability. A pricing indication may therefore differ from both statutory and incremental costs while remaining coherent when the objective is disclosed.

An insurer may allocate shared information technology expense equally per policy, in proportion to time spent servicing, or according to an agreed capacity metric. The methods produce different policy-level results even when all three reconcile to the same organization-wide cost. This allocative dependence is a material result: a premium's expense content can reflect managerial or regulatory allocation choices as well as observed resource consumption. Treating the allocation rule as if it were directly observed cost would obscure this element of judgment.

Expense recognition versus pricing

The accounting date of an acquisition payment and the prospective recovery of that cost through premiums serve different objectives. Under US statutory insurance accounting, NAIC SSAP 71 acquisition-cost treatment requires policy acquisition costs and commissions to be expensed as incurred. The NAIC's Financial Analysis Handbook describes the distinction from US GAAP's capitalization and amortization of qualifying deferred acquisition costs. The statutory expense date alone does not determine the economically suitable lifetime allocation of acquisition costs in a pricing calculation.

IFRS 17 Insurance Contracts measures groups of insurance contracts and recognizes insurance service results in relation to coverage. Its insurance acquisition cash-flow accounting considers directly attributable cash flows and their allocation, with specific requirements and optional simplifications under the premium allocation approach. The IASB's acquisition-cash-flow deliberations describe systematic and rational allocation for relevant acquisition flows and expected future groups. Neither the IFRS accounting policy nor its deferral mechanics can be substituted mechanically for a prospective rate indication; the price model must still recover expected economic cash flows across the actual policy and renewal horizon.

Similarly, Solvency II Article 31 requires projection of administrative, investment-management, claims-management, and acquisition expenses in the best estimate and an objective, consistent allocation of related overhead. This addresses the valuation of obligations, including servicing expenses, rather than a direct instruction to charge a particular premium. The cross-system lesson is precise: financial recognition, regulatory liability measurement, and competitive product pricing use different time frames and intended measures even when they begin from the same underlying expense.

Commission and Acquisition Expenses

Contractual forms and distribution functions

A commission transfers compensation to an intermediary or another party involved in acquiring or servicing insurance business. It may be calculated on written or earned premium, net or gross premium, an amount of new business, retained business, persistency, profitability, or other contractually defined outcomes. The definition, premium base, premium adjustments, clawbacks, cancellations, and payment timing determine the realized cost. A nominal rate of 12% is incomplete until the contractual base and conditions are specified.

Intermediaries can perform search, matching, risk explanation, submission preparation, underwriting information collection, placement, servicing, and claims advocacy. Cummins and Doherty (2006) analyze these functions and the economics of commercial property/casualty distribution in The Economics of Insurance Intermediaries. They describe both premium-based and contingent compensation and discuss information asymmetries and market-making roles. Their empirical findings pertain to the studied market and period and cannot be read as a universal estimate of intermediary value or of commission pass-through in every insurance class.

Distribution-system comparisons have long recognized organizational differences. Barrese and Nelson (1992), in their agency cost comparison, examine independent and exclusive agency expense differentials. Regan (1997) analyzes vertical integration and transaction costs, explaining why ownership of customer relationships and transaction-specific information can alter the economic merits of different distribution structures. Expense ratios alone do not establish causation: risk complexity, claims experience, service intensity, market selection, reporting arrangements and the transfer of functions between insurer and distributor must also be considered.

Gross-premium mathematics and commission incidence

Let L denote expected losses and any loss adjustment expense included in the loss provision, F the expected per-policy costs that do not vary directly with premium, c the commission rate charged against the modeled premium, v other premium-proportional expenses, t premium-based taxes and assessments, and m an underwriting-profit or contingency target represented here as a proportion of premium. Under a single-period, undiscounted, homogeneous, fully collectible premium assumption,

P=L+F+(c+v+t+m)P,P=L+F1−c−v−t−m.P=L+F+(c+v+t+m)P, \qquad P=\frac{L+F}{1-c-v-t-m}.

This is the conventional premium gross-up, consistent with the CAS Basic Ratemaking treatment. A positive finite premium requires 1 - c - v - t - m > 0. The formula is an accounting identity under the assumptions, not a guarantee that customers will accept the resulting price. It also does not price the time value of money, risk-dependent taxes, premium refunds, acquisition costs payable on a different base, or multi-year retention unless explicitly extended.

The economic effect of a commission rate is nonlinear because the commission is assessed on a premium that includes its own recovery. Holding L, F and the other proportions constant, the sensitivity is

∂P∂c=L+F(1−c−v−t−m)2>0,∂ln⁡P∂c=11−c−v−t−m.\frac{\partial P}{\partial c}=\frac{L+F}{(1-c-v-t-m)^2}>0, \qquad \frac{\partial\ln P}{\partial c}=\frac{1}{1-c-v-t-m}.

A one-percentage-point change in commission therefore raises the price by more than one percentage point whenever the denominator is below one. The commission share is an accounting fraction cP, while the difference in premium between two compensation systems also changes the amounts collected for all other percentage-based provisions. Calling c the entire customer price effect understates the feedback built into the gross-up.

Take a fully hypothetical contract with L = $600, F = $50, commission c = 12%, other premium-variable expense v = 6%, premium tax t = 2%, and margin m = 5%. The denominator is 0.75 and the indicated premium is $866.67, comprising expected losses $600.00, fixed expenses $50.00, commission $104.00, other variable expenses $52.00, premium tax $17.33, and margin $43.33, subject to rounding. Raising c to 15% produces a denominator of 0.72 and P = $902.78. All quantities are constructed for this monograph, and the profit component is separated from the ASOP 29 expense definition.

When a commission is levied on a narrower base B(P), the equation becomes P = L + F + vP + tP + mP + cB(P). If B is net of an explicit fee or excludes a premium component, direct use of the gross-premium percentage c in the preceding denominator is incorrect. If compensation is fixed at k dollars per policy, the equation is P = (L + F + k)/(1 - v - t - m), with no c term. Mixtures of fixed and percentage remuneration can be solved with both terms; contingent payments require an expected-value allowance reflecting their contractual triggers and correlation with loss and premium outcomes.

New business, renewal and contingent compensation

Acquisition expense often occurs at contract inception while service costs accrue during coverage. A broker or agent may receive a higher new-business rate than renewal rate, or compensation may depend on retention and loss experience. A single period commission ratio conceals these distinctions. A multi-period pricing assessment may use discounted expected revenue and cost, conditioned on cancellation, renewal, commissions, claims, expense inflation and investment assumptions:

∑k=0TdkskPk=∑k=0Tdksk{Lk+Fk+vkPk+ckPk+tkPk}+K0+M,\sum_{k=0}^{T} d_k s_k P_k =\sum_{k=0}^{T}d_k s_k\{L_k+F_k+v_kP_k+c_kP_k+t_kP_k\}+K_0+M,

where d is a discount factor, s a persistency or exposure survival probability, K_0 initial acquisition outlay excluded from the recurring F terms, and M the appropriately defined discounted surplus or margin target. This equation is a cash-flow accounting construction; it is not a recommendation to subsidize an inadequately priced initial term with speculative renewals. The contractual and legal availability of future renewals, renewal price restrictions, lapse selection, and recoverability of acquisition outlays determine how such a calculation can inform present pricing.

Contingent commissions based on low loss ratios or high business volume create both cost uncertainty and intermediary incentives. A rational pricing allowance should use the expected payment across relevant outcomes, incorporating any contract caps and nonlinearity. For a bonus equal to b times eligible premium if the portfolio loss ratio is below threshold h, expected bonus equals b E[P_eligible 1{loss ratio < h}], not bE[P_eligible] times an assumed independent trigger probability unless independence is justified. Its correlation with favorable claims experience may reduce net risk in some arrangements; at the same time, the bonus could incentivize selective placement or service choices. The empirical and theoretical treatment of both possibilities in Cummins and Doherty (2006) supports evaluation of incentives along with cost.

Channel selection and incentive effects

Direct distribution may reduce commissions while raising the insurer's own advertising, lead acquisition, contact-center, customer support or digital infrastructure costs. Agency distribution may transfer those functions outside the insurer, so comparisons must restate the scope of activities on a common basis. Higher distribution compensation can purchase customer access, information quality, retention and service. The economically relevant comparator is the change in total expected cost and value for the contract mix produced by the channel, not the commission line alone.

Consider two channels for the synthetic policy just described. Replacing the 12% broker commission with 4% direct-channel commission, while holding L + F = $650 and all other proportional components totaling 13%, lowers indicated premium from $866.67 to $650/0.83 = $783.13. The $83.53 difference is a modeled premium difference, not a claim that direct distribution is intrinsically cheaper. If the broker channel instead lowers expected losses and fixed servicing costs together by approximately $62.65, its premium becomes comparable to the direct price under the stipulated denominator. A comparison that ignores such differences confuses a compensation rate with distribution-system efficiency. It also assumes identical policies; an actual insurer would need channel-specific risk adjustment and retention analysis.

Fixed and Variable Expense Loads

Estimating cost drivers

Each expense pool should be assigned to the driver that best reflects the way the cost arises, subject to observable data and the decision purpose. Premium taxes may fit a premium base; policy issuance and annual notices may fit policy counts; inspections and complex underwriting may fit time or case mix; claim investigation may fit claim volume and complexity; payment processing may fit transactions or premium collections. An allocation that is administratively convenient but causally weak can make large policies absorb excess overhead while small policies are undercharged for servicing, or reverse those effects.

For a budgeted operating pool C and expected driver volume D, a simple unit provision u = C/D assigns e_i = u d_i to policy or segment i. This is transparent when C and D belong to the same horizon and the driver is meaningful. It is sensitive to projected volume: with annual servicing capacity cost of $1,000,000, an expected 10,000 policies yields $100 per policy; at 8,000 policies the same budget yields $125, and at 12,000 it yields $83.33. The per-policy provision depends materially on volume assumptions even when the total fixed cost is unchanged. Forecast errors therefore change the realization of both rate adequacy and expense ratios.

Time-driven activity-based costing estimates the practical capacity cost rate of a resource and assigns it according to activity time and complexity. The operational approach developed by Kaplan and Anderson (2004) in Time-Driven Activity-Based Costing provides one conceptual framework, although its general management-accounting model does not constitute an insurance-specific actuarial standard. For an illustrative underwriting pool costing $960,000 annually with 32,000 practical labor hours, the rate is $30 per hour. A case requiring 18 minutes of underwriting capacity consumes $9 at that rate; a complex case requiring 75 minutes consumes $37.50. Idle capacity, shared infrastructure and changes in process design must still be reconciled to the budget; measured labor minutes are not a universal proxy for all overhead.

A complete cost ledger can distinguish capacity supplied, capacity used, and reserve capacity. Loading all current idle capacity onto active policies creates elevated average costs when volume is temporarily low. Ignoring the cost of capacity needed to serve future business underprices the expected operation. For pricing, the relevant question is which capacity the insurer expects to maintain at the volume contemplated by the rate. The cost of excess capacity may be treated differently in management performance reporting; the allocation choice should be documented.

Volume, trend and operational changes

Historical expenses require consistent exposure and inflation treatment. A salary-intensive cost pool may follow compensation growth; a cloud-processing pool may change with transaction volume or unit technology cost; premium taxes generally change with premium and law; commissions change with contract terms and distribution mix. Applying claim severity trend to all operating expenses is conceptually unjustified. ASOP 13 on trending distinguishes projections between exposure periods from development within an exposure period and calls for trends appropriate to their purpose, data and relevant influences. ASOP 29's expense trending also recognizes that expense trend may differ from loss or premium trend.

For pool j, let historic cost C_j0 be adjusted for classification and one-time items; then a prospective illustration is C_j1 = C_j0(1+g_j)^h + ΔC_j, with g_j the relevant annual trend over horizon h and ΔC_j explicit operational, legal or capacity adjustments. If staffing changes stepwise, a smooth multiplicative trend alone will miss the discontinuity. If a new platform reduces per-transaction labor but introduces recurring license charges, the two components must be modeled separately. Double counting can arise when an operational change is included both in trend selection and in ΔC_j.

Commission forecasting should begin with contract terms and expected portfolio distribution rather than assume historical commission-to-premium ratio remains constant. The correct portfolio rate can be a weighted contractual measure, with the weights corresponding to the compensation base. An arithmetic average of channel commission percentages weighted by policy count is generally incorrect when premiums vary widely across channels; a premium-weighted average may still be misleading when the premium itself responds to the commission structure or when written and earned exposures differ.

Statistical and activity-based methods

For repeated transaction costs, a generalized linear model can relate expected cost to coverage type, policy size, underwriting complexity, channel, processing mode, geography and exposure. If the dependent variable is expense dollars, link, distribution, zero handling, offsets and exposure duration must match the observed cost-generating process; count models and severity models may be appropriate for different activities. The CAS Generalized Linear Models monograph by Goldburd, Khare, Tevet and Guller (2025 revision) explains practical insurance-rating model development, but its availability does not validate a proposed expense model by itself. The model requires actual operational expense observations, not merely claim outcomes or premiums.

The quantity of data can support detailed classification without resolving common-cost causation. A statistical association between premium and expense may be induced because high-premium policies receive more attention or because pricing already incorporates a percentage loading. Treating such endogenous correlation as proof of marginal cost would create a circular allocation. A defensible analysis compares premium-based, policy-based and activity-based projections on holdout periods; checks cost totals against the ledger; assesses stability by class and channel; and evaluates residuals where policies have similar premium but different operational complexity. Where data are scarce, grouped cost studies and expert-supported process estimates can be more transparent than a fragile high-dimensional model.

The ASOP 23 data-quality standard is relevant to selection and review of source data, while ASOP 56 on modeling addresses model purpose, design, validation and communication within its professional scope. A proper data dictionary should preserve gross versus net premium, written versus earned premium, policy inception and reporting periods, expense posting timing, commission accrual and reversals, intermediary identities, multi-party distribution chains, and any changes to systems or cost centers. Reconciliation of the total expense pool to audited or controlled accounting data is a stronger first test than predictive fit alone.

Controls, governance and uncertainty

The governing control should be a complete trace: cost pool → source ledger and operational process → assigned driver → forecast volume → prospective cost allowance → rate class or contract → sensitivity and reasonableness check. Each arrow represents a testable assumption or a documented allocation choice. When responsibilities are outsourced, the assessment should combine payments to the contractor with costs retained by the insurer, while distinguishing a true transfer of activity from an apparent accounting substitution.

Four uncertainty types deserve explicit review. Parameter uncertainty concerns cost rates, trend coefficients and forecast volumes. Structural uncertainty concerns changes in underwriting processes, distribution contracts and technology. Allocation uncertainty concerns the justification for mapping shared costs to policy classes. Behavioral uncertainty concerns policyholder demand, intermediary placement, renewal and the loss mix induced by prices. A cost model with excellent historic statistical fit may remain poor at projecting structural changes. Sensitivity intervals are most useful when they show which assumptions actually move the indicated premium and expose breaks in the cost-driver relationship.

Exposure and Premium-Based Allocation

Alternative allocation rules

The all-variable expense method applies one or more selected expense ratios to premium. It can approximate recurring premium-proportional taxes and commissions. The premium-based projection method begins with total historical expense and premium ratios, adjusted to the prospective period. The exposure- or policy-based method estimates a cost per exposure, coverage or policy, while more differentiated methods allocate by actual activity. These alternatives, their assumptions and their limitations are discussed in Werner and Modlin's ratemaking expense chapter. Their comparability requires the same cost universe and forecast horizon.

For insurer portfolio K, let direct policy-level cost be D_i, variable cost rate r_i, and common expense pool F_K. Under a full-cost allocation with nonnegative weights w_i summing to one, expected provision for policy i may be represented as E_i = D_i + r_i P_i + w_i F_K. Policy count allocation uses w_i = 1/N, exposure-weighted allocation w_i = x_i/Σx, and premium-weighted allocation w_i = P_i/ΣP. Premium weights create an endogenous system when price itself includes expense loading; an iterative or simultaneous solution is appropriate where the effect is material. All three can reconcile the portfolio total while distributing expense unequally across policies.

Full cost attribution is separate from marginal economic cost. If another small policy can be issued without increasing platform cost, its incremental shared fixed cost may be near zero even though a share of the platform must be covered by the portfolio over time. A competitive premium based solely on incremental cost can be a rational short-run decision if an explicit financing strategy exists, but a persistent portfolio priced below recurring total cost lacks self-financing adequacy. Conversely, assigning a large idle-capacity charge to a newly acquired policy may make otherwise efficient growth appear unprofitable. The correct method follows the decision horizon, with common-cost recovery made explicit.

Worked premium and elasticity calculations

Reconsider the illustrative base P = (600 + 50)/(1 − 0.12 − 0.06 − 0.02 − 0.05) = $866.67. Table 1 makes the commission and fixed-cost mechanics directly auditable.

Hypothetical case Loss + fixed cost Commission Other variable + tax + margin Denominator Indicated premium
Base $650 12% 13% 0.75 $866.67
Commission 15% $650 15% 13% 0.72 $902.78
Commission 8% $650 8% 13% 0.79 $822.78
Fixed cost +$20 $670 12% 13% 0.75 $893.33
Expected losses +10% $710 12% 13% 0.75 $946.67
Direct channel 4% $650 4% 13% 0.83 $783.13

The 10% loss stress raises L from $600 to $660; all other assumptions remain fixed. The 8% commission stress uses 1 − .08 − .13 = .79. The table isolates one assumption at a time, rather than presenting an empirical range of observed premiums. Under the fixed-cost scenario, the $20 increase becomes a $26.67 increase in gross premium because every dollar of fixed expense also requires financing the associated proportionate items.

The local elasticity of premium with respect to c depends on the definition of the change. In this additive-rate formulation, ∂P/∂c = (L+F)/d² where d is the denominator, so at d = .75 it equals $1,155.56 per unit change in c. A one-percentage-point change in commission therefore yields a first-order premium change of about $11.56, with the exact finite change slightly different. The term “one percentage point” here means Δc = .01, not a 1% proportional rise in the existing commission rate. Explicit units avoid a common source of pricing error.

Policy-size and portfolio cross-subsidies

Suppose two policies incur the same $80 fixed administration cost, but their expected loss costs are $200 and $2,000. Let all premium-proportional expenses and margin sum to 18%. The small policy's indicated premium is (200+80)/.82 = $341.46; the large policy's is (2000+80)/.82 = $2,536.59. The $80 fixed expense is approximately 23.43% of the small policy's premium but 3.15% of the large policy's premium. Charging a uniform expense percentage instead would allocate more absolute cost to the larger contract even though the stipulated administration task costs the same. Whether a percentage convention is commercially or regulatorily acceptable is a separate question; its distributional effect is mathematically observable.

A uniform per-policy amount is not necessarily equitable either. If the large policy requires multiple risk-engineering surveys and specialized claims operations, its direct activity costs may be greater; if small policies produce disproportionately high billing and customer-service transactions, the reverse may occur. The scientifically appropriate response is a driver study, sensitivity assessment and explanation of any purposeful pooling of common costs. A choice to cross-subsidize for accessibility or pooling can be described explicitly; it should not be disguised as a causal cost estimate.

Expense flattening and minimum premiums illustrate the tension. A lower percentage and fixed charge can reduce absolute charges on high-premium risks, while a minimum premium can secure recovery of policy issuance costs on small risks. Both policies affect the price schedule and therefore demand. A minimum premium above the actuarial indicated risk premium can change which customers purchase coverage, altering the average underlying loss cost of the remaining portfolio. The overall result is a coupled demand–cost problem that must be examined with actual take-up and exposure data.

Channel, renewal and reinsurance illustrations

A mixture of distribution channels should be projected by the business actually expected, not by an unadjusted historical average. If a channel's share changes, its acquisition costs, commission base, policy size, persistency and risk mix may all change. Define segment-level premiums P_j = (L_j+F_j)/(1-r_j), with r_j its own contractually grounded proportional costs. Aggregate premium is ΣN_jP_j only if N_j corresponds to the units represented by the P_j and there is no omitted interaction. Aggregating at a single weighted average rate generally loses heterogeneity when numerators, premium bases and denominators differ.

For acquisition spending paid upfront, expected initial economics should be evaluated alongside renewal outcomes while keeping each term's timing intact. If a $150 initial acquisition payment is expected to support three annual contract periods, dividing it by three mechanically assumes equal realized exposure and full persistency. With lapse probabilities or unequal annual premiums, the effective per-renewal exposure share changes. Discounted expected contract cash flows and rules on recoverability provide a better basis. Financial recognition under IFRS 17 remains a distinct reporting task; pricing may take a different view of the horizon and the probability of future business.

For reinsurance, suppose a cedant pays gross reinsurance cost R, expects allowance A, loss recoveries V and any contingent profit share H, all measured consistently over the same period. An illustrative expected net reinsurance cash cost is R − A − V − E[H], with reinstatement and timing adjustments when relevant. Whether this belongs in expense provisions or another risk-cost component depends on the pricing architecture and applicable professional framework. The cedant must also model its retained loss, so subtracting recoveries from retained losses and again from a net reinsurance expense would count the benefit twice. ASOP 29 on reinsurance identifies the relevant reinsurance payments, allowances and contract contingencies but does not require one universal bookkeeping format.

Rate Indication Effects

From cost analysis to indicated rates

The rate indication converts a prospective cost estimate into a comparison with the insurer's current rate level. It is necessarily prospective because the policies to be issued may differ from the experience population. CAS ratemaking principles are a historical reference for regulated US ratemaking; the CAS statement was originally published in 1988, rescinded in 2020, and reinstated in 2021 as a reference, which should not be confused with current law or an actuarial standard. Current professional guidance includes ASOP 53's prospective framework.

With projected loss and fixed cost A, and total permissible premium-proportion r < 1, indicated premium equals A/(1−r). For current average premium P_0, the indicated rate-level change is I = A/[(1−r)P_0] − 1. If A = $650, r = 25%, and P_0 = $800, the indicated premium is $866.67, and I = 8.333%. The current rate may generate different observed expense ratios than the prospective rate, so inserting a percentage from historical premium into r requires projecting the contract terms and cost behavior. The correct denominator is a prospective allowance, not an unexplained residual needed to achieve a preferred rate change.

When business volume depends on the new rate, total fixed cost F_total and policy count N(P) are jointly relevant. An illustrative full-cost relation is P = [L(P)+F_total/N(P)+A(P)]/(1−r(P)). This can have multiple or unstable solutions under demand response, especially when a steep increase in price drives away low-cost customers or reduces policy volume. The equation clarifies why a per-policy provision derived from an assumed fixed N should be stress-tested. A ratemaking analysis must account for the loss experience and risk mix expected at the new rate rather than treating demand as irrelevant.

Numerical sensitivity and stress testing

Table 2 illustrates a volume assumption under a fixed annual operating pool without changes to loss cost or the proportional loading.

Expected policies Fixed annual cost Fixed cost per policy Pure loss per policy Other share of gross premium Indicated premium
8,000 $1,000,000 $125.00 $600 25% $966.67
10,000 $1,000,000 $100.00 $600 25% $933.33
12,000 $1,000,000 $83.33 $600 25% $911.11

The example holds the operating cost pool fixed and presumes every policy has the same risk and exposure; it is a sensitivity diagnostic only. A production insurer would test correlated changes in premium level, volume, expected losses, exposure mix and cost staffing. In an adverse scenario, lower volume increases the fixed allowance precisely when the insurer faces pressure to raise its rate, creating a potential reinforcing effect. In an efficient growth scenario, additional capacity utilization can lower per-unit costs, provided service quality and risk-selection discipline remain adequate.

A further diagnostic separates expected-versus-actual variances into price or wage variance, volume variance, productivity variance, mix variance, and allocation variance. A rise in the expense ratio may reflect declining premium, a temporary volume reduction, an actual efficiency loss, or the movement of an activity previously undertaken by a broker into the insurer's accounts. Only the decomposition identifies the corrective action. Performance measurement should not automatically incentivize reducing claims-handling capacity or compliant advisory services simply because their ledger cost rises.

Prospective, accounting and prudential boundaries

The same economic expense can be recognized at different times under US statutory accounting, US GAAP, and IFRS 17, and can enter a solvency best estimate on yet another basis. The NAIC financial analysis guidance explains statutory immediate expensing of acquisition costs, whereas IFRS 17's contract-group measurement recognizes insurance-contract cash flows and service performance in its reporting architecture. The finance team and pricing actuary should reconcile these outcomes while preserving their distinct purposes and time horizons. An apparent loss on an accounting inception date does not by itself establish that a multi-period rate is inadequate; conversely, delayed financial recognition does not make a policy economically profitable.

Under the European Solvency II framework, the Article 31 expense projection includes expenses required to service recognized insurance and reinsurance obligations and prescribes realistic, objective, consistent overhead allocation in best estimates. It is not equivalent to an insurer's acquisition strategy or expense-inclusive quotation. The prudential requirement to value existing liabilities should be distinguished from the price to attract future policyholders, even where certain future-business assumptions inform the overhead projection.

The 2025 revision of ASOP 20 became effective June 1, 2026 for work within its scope and concerns analysis of property/casualty cash flows including discounting. It provides a relevant temporal qualification where a price exercise projects expense cash flows and their timing; ASOP 29 remains the directly specialized source on expense provisions. Time value, interest rates, payment timing and risk-transfer structure can affect which undiscounted simplifications remain useful.

International application and limits

Legal treatment of commissions differs across product types, markets and distribution arrangements. The European Union's Insurance Distribution Directive requires distributors to act in customers' best interests and identifies remuneration transparency and conflict-of-interest obligations. Article 19 addresses disclosure of the nature and basis of intermediary remuneration, including fees, commissions and other economic benefits, subject to the directive's scope and national implementation. It does not establish a universal commission cap or uniform premium expense percentage. Nor does a disclosure requirement settle whether a chosen cost allocation is economically accurate.

The United Kingdom's FCA product governance rules require relevant non-investment insurance products to demonstrate fair value, including consideration of distribution remuneration and cumulative distribution chains. The rules address the total price, benefits and distribution value; they do not prescribe that every commission must be eliminated. These obligations should be applied only to covered firms and products and the rule version appropriate to the period being analyzed. For international comparison, the analytical categories are portable, while legal duties and expense constraints require local confirmation.

Transparency and Efficiency

Remuneration, conflicts and product value

A distribution commission may pay for real services and may at the same time create an incentive to place business with a particular insurer, increase the amount of business, or prefer certain policies. The economically relevant assessment examines contract design, actual services, customer outcomes, risk information, and any incentive conflict. Under the IDD best-interest obligation, distributor remuneration and sales incentives must not undermine the duty to act in the customer's best interests. The FCA's fair-value framework requires oversight of distribution arrangements and remuneration's effect on the customer's total price and product value within its applicable scope.

Remuneration transparency and actuarial expense adequacy are complementary governance controls. A rate can accurately recover a commission and still generate questionable value if distribution activity produces little customer benefit. A distribution product can provide meaningful advice but remain mispriced if the commission and servicing costs are omitted from the rate. Efficiency is therefore measured by the services and outcomes delivered per unit of economic resource consumed, with risk and product complexity held appropriately comparable.

The scholarly evidence discourages simple rankings of independent agents, exclusive agents, and direct channels based solely on the reported commission ratio. Barrese and Nelson's cost study and Regan's transaction-cost analysis belong to different institutional and historic settings. Their implications are conditional on organization, product complexity, customer relationships and information flows. Cummins and Doherty's intermediary analysis shows why market-making, information and contract incentives belong in the same analysis as expense pass-through. None of these studies supplies a single contemporary universal commission rate.

Allocation, fairness and competition

Every common-cost allocation generates an incidence pattern. Premium-weighted recovery allocates more to higher-premium contracts; policy-count recovery allocates equal dollars to contracts even if their insurance values differ; exposure and activity allocation attach costs to specified drivers. These methods affect relative prices, demand, product access and competitive positioning. Under a uniform administrative cost, a pure premium percentage allocates more expense to a high-risk policy even if it is no more costly to administer. Under genuinely complexity-sensitive processing, a flat fee may instead make simple policies contribute to the servicing of complex risks. The correct normative judgment requires an articulated cost objective and evidence about the service process.

Risk classification and expense classification also interact. If an observed operational variable proxies for legally protected or socially sensitive customer traits, a technically predictive model may raise questions about permissible or fair product design. Conversely, equalizing a cost driver without acknowledging different real services can make a system economically difficult to sustain. The present dossier establishes the empirical decomposition needed for these judgments; it does not create a universal fairness rule or a new principle of allocation. Appropriate restrictions are determined within each legal and product setting.

Competition makes commission loadings particularly consequential. When all premiums rise to recover higher acquisition costs, demand may contract, which reduces volume and raises fixed per-policy costs. When an intermediary reduces information asymmetry or facilitates better matching, its commission may lower the loss or customer acquisition costs needed elsewhere in the equation. A price comparison without portfolio selection and service quality can therefore reverse the apparent efficiency ordering. The research implication is to observe gross and net customer prices, total channel costs, claims experience, persistency, and service quality together.

Operational diagnostics and analytical synthesis

An auditable expense study should answer five concrete questions. What expense was actually incurred and for what function? Which dimension of business volume or complexity generates it? How will that dimension change in the prospective pricing period? Which contractual or regulatory constraints govern its recovery? How sensitive is the indicated premium to alternative plausible drivers and forecasts? The sequence leads naturally to a reconciled ledger, expense data dictionary, driver map, forecast scenarios, premium equation and transparent editorial explanation of methodological judgment.

The original synthesis of this monograph has four results. First, expense adequacy and allocation accuracy are distinct: an aggregate price can recover total expenses while materially misallocating them among policies, channels or classes. Second, a commission percentage enters the premium equation through a denominator, so the premium impact of contractual remuneration depends on all other proportional provisions and on the compensation base. Third, fixed expense and policy demand can be jointly determined in the forward pricing period; volume sensitivity is thus part of expense adequacy, not merely an accounting afterthought. Fourth, intermediary compensation should be assessed on a common-function basis with risk selection, service value, and customer outcomes, while maintaining the different measurement objects of pricing, accounting and prudential reporting.

These results are analytical inferences from established methods, primary standards, and transparent constructed examples. Their novelty is the organization and comparative integration of those elements in the present monograph, rather than a claim to have originated the premium gross-up, activity-based costing, or the economics of insurance intermediation. The central empirical limitation is deliberate: no anonymized insurer ledger, commission contract portfolio, policy-level risk file, or controlled causal study is asserted. For any actual insurer, the quantitative findings would require calibration against those data, local law, reporting bases and observed business response.

Conclusion

Insurance expense provisions are prospective claims on operating resources, and their allocation shapes both total premium adequacy and the distribution of insurance costs across customers and risks. Reliable expense analysis starts with a complete, non-overlapping cost taxonomy; identifies premium, policy, exposure, claim and activity drivers; distinguishes direct and common costs; adjusts historical observations for future conditions; and reconciles the resulting projections with the actual contract and business plan. Current US actuarial standards supply a disciplined professional reference for that work. International accounting, solvency, and distribution rules establish further obligations whose objectives are related but distinct.

Commission loadings warrant special attention because they can depend on premium itself, vary between new and renewal contracts, be conditional on portfolio outcomes, and alter incentives in risk selection and customer service. Their gross-premium effect is nonlinear, and comparative channel efficiency cannot be inferred from a commission ratio alone. Fixed costs create a second important feedback: a decline in policy volume raises full-cost expense per policy even without any change in the underlying annual budget. The numerical cases make both mechanisms reproducible and demonstrate the allocation effects that a purely proportional method may conceal.

A defensible indicated rate is therefore the outcome of an explicitly defined prospective cost calculation, a justified allocation policy, appropriate uncertainty and time assumptions, and a consistent treatment of commissions and other percentage provisions. Where legal or fairness considerations constrain the indicated rate, the economic consequence and any financing gap should be made visible. The final scientific judgment is that the accuracy of an expense loading depends on the correspondence among the expense's underlying economic behavior, the unit used to allocate it, and the purpose and horizon of the price. That correspondence is the operational foundation of credible insurance pricing expense allocation.

Appendix: Reproducible calculations

All examples are deterministic and synthetic. Currency is illustrative US dollars; no empirical insurer observations were used. The common gross-premium function is premium(loss, fixed, commission, other) = (loss + fixed) / (1 - commission - other) where other includes specifically named percentage-based expense, tax, and margin allowances, and the denominator must be positive. Rounding to two decimals occurs after full-precision calculation.

Quantity Exact operation Result rounded
Base premium 650 / 0.75 $866.67
Base commission 0.12 × (650 / 0.75) $104.00
Base other variable 0.06 × (650 / 0.75) $52.00
Base premium tax 0.02 × (650 / 0.75) $17.33
Base margin 0.05 × (650 / 0.75) $43.33
15% commission premium 650 / 0.72 $902.78
8% commission premium 650 / 0.79 $822.78
Fixed cost plus $20 670 / 0.75 $893.33
10% loss stress (660 + 50) / 0.75 $946.67
Direct-channel premium 650 / 0.83 $783.13
Broker–direct price gap 650 / 0.75 − 650 / 0.83 $83.53
Breakeven broker cost saving 650 − 0.75 × (650 / 0.83) $62.65
Small-policy premium 280 / 0.82 $341.46
Large-policy premium 2080 / 0.82 $2,536.59
Fixed cost at 8,000 policies 1,000,000 / 8,000 $125.00
Fixed cost at 10,000 policies 1,000,000 / 10,000 $100.00
Fixed cost at 12,000 policies 1,000,000 / 12,000 $83.33
Underwriting hourly capacity 960,000 / 32,000 $30.00
Simple-case underwriting 30 × 18 / 60 $9.00
Complex-case underwriting 30 × 75 / 60 $37.50
Rate indication versus $800 (650 / 0.75) / 800 − 1 8.33%

For the fully specified general expression, expected loss and loss adjustment expense must be mutually exclusive of the expense provision; fixed and variable expenditure must be projected for the same future exposure units; the commission base must be consistent with the percentage applied; and the margin's definition must be separated from the operating expense categories. These consistency checks are part of the mathematical meaning of each result.

References