Methodology

Under 26 U.S.C. § 162(a)(1), businesses may deduct “a reasonable allowance for salaries or other compensation for personal services actually rendered.” For S-Corporation owner-employees, reasonableness is judged by what would ordinarily be paid for like services by like enterprises under like circumstances (Treasury Regulation § 1.162-7(b)(3)) — in practical terms, what it would cost to hire someone else to perform the work the owner does.

S-Corporations pass through income to shareholders, who pay income tax but not self-employment tax on distributions. The IRS requires that owner-employees pay themselves reasonable compensation as W-2 wages — subject to FICA taxes (Social Security and Medicare) — before taking distributions. When compensation is set unreasonably low, the IRS may reclassify distributions as wages, assess back payroll taxes, and impose penalties and interest.

There is no statutory formula for “reasonable.” The determination is inherently factual, based on the nature of the services performed, the owner’s qualifications, and comparable market data. The IRS and the Tax Court have established analytical frameworks for this determination, which this tool implements using publicly available wage data.

Courts and the IRS evaluate reasonable compensation through a multi-factor analysis rather than a single test. The nine factors below are the list the IRS summarizes in Fact Sheet FS-2008-25. The multi-factor approach itself traces to early appellate cases such as Mayson Mfg. Co. v. Commissioner, 178 F.2d 115 (6th Cir. 1949), which held that reasonableness turns on the whole situation, with no single factor decisive. The IRS’s own nine-factor list appears separately in Fact Sheet FS-2008-25, which says the courts that have ruled on the issue based their determinations on the facts and circumstances of each case.

The Cost Approach decomposes the owner’s role into the specific business functions performed and prices each function independently using BLS wage data at the appropriate skill level and geographic location. The IRS Job Aid refers to this approach as the Cost Approach: the cost to replace the owner’s services by hiring separate individuals to perform each function.

This approach fits small S-Corporation owners who perform multiple business functions. Most small business owners divide their time across administrative, operational, financial, and specialized tasks — no single occupation classification captures the full scope of their work. The Cost Approach accounts for this by valuing each function at its own market rate rather than forcing a single job title onto a multi-function role.

The Market Approach compares the owner’s overall position to a single comparable occupation in their geographic area. Rather than decomposing the role into tasks, it identifies the primary occupation that best describes the owner’s function and applies the BLS wage for that occupation at the owner’s experience level.

The two approaches can produce different figures, and neither reliably runs higher than the other: the Cost Approach blends the wages of every function the owner performs (including lower-wage operational tasks), while the Market Approach prices the single closest comparable role. Running both frames a range to evaluate — the Cost Approach represents the replacement cost of the owner’s actual work mix, while the Market Approach represents what a single comparable professional would earn.

The IRS Job Aid notes that “the reconciliation in the case of Reasonable Compensation will generally rest heavily on the market approach.” Each report applies one approach; the owner and their tax professional can run the analysis under more than one of these approaches and compare the results to choose the figure most appropriate to their circumstances.

The Income Approach asks whether a hypothetical independent investor would consider the owner’s compensation reasonable given the return generated by the business. Rather than pricing the owner’s labor directly, it works backward from business performance: if the company’s return on equity exceeds what an outside investor would expect, the owner’s salary may be too low.

The independent-investor framing was articulated in Elliotts, Inc. v. Commissioner, 716 F.2d 1241 (9th Cir. 1983) and prominently adopted in Exacto Spring Corp. v. Commissioner, 196 F.3d 833 (7th Cir. 1999), where the Seventh Circuit reasoned that when “the investors in his company are obtaining a far higher return than they had any reason to expect, [the owner’s] salary is presumptively reasonable.” Elliotts and Exacto were excessive-compensation cases — the question there was whether pay was too high — but the same investor-return logic applies symmetrically to test whether an owner’s salary is too low.

This approach is most useful when comparable wage data is unavailable for the owner’s role, or when the owner drives outsized returns that are not fully captured by labor-market comparisons. It provides a complementary data point alongside the Cost and Market Approaches.

This mapping anchors the wage percentile to a documentable, verifiable fact — years of experience performing that specific function — rather than a subjective self-assessment of skill level. “I have 4 years of bookkeeping experience” is more readily substantiated than “I am an average bookkeeper.”

The proficiency level is set independently for each task. An owner might have 15+ years of experience in their core professional function (Expert, 90th percentile) while having only 2 years of experience managing employees (Developing, 25th percentile). The Cost Approach captures this nuance.

Note: The experience-year thresholds defining each level are informed judgments designed to provide reasonable anchors, not empirically derived cutoffs. They are intended to produce a more documentable basis for percentile selection than subjective self-assessment.

Percentile suppression: BLS does not publish a wage for every percentile of every occupation and area. A percentile can be withheld, in which case BLS releases no estimate at all: its own FAQ says estimates “may be withheld from publication for a number of reasons, including failure to meet BLS quality standards or the need to protect the confidentiality of our survey respondents”, and that OEWS “is not able to provide the specific reason that an estimate was not released”.

A percentile BLS withheld is handled differently, and not by moving to a wider geography. Where a median was published for the same occupation and area, the median is used in its place and the substitution is stated; across the BLS releases this report can draw on, no row has met that condition, because for a given occupation and area the hourly percentiles have been published together or withheld together, leaving a withheld percentile with no usable wage on its row. Geography changes only when the occupation has no usable row in the local wage area at all, and only after the prior year’s release has been tried for whichever area BLS assigned the county in that release; the report then works outward from the county’s own BLS wage area to the state and then to national, stopping at the first level that publishes the occupation. Every substitution the report actually made is disclosed.

Annual hours are capped at 2,080 — the standard full-time equivalent of 40 hours per week for 52 weeks. This matches how BLS wage data is structured and the standard full-time-year basis used in reasonable-compensation analyses, including by the IRS expert in McAlary (below).

In Sean McAlary Ltd., Inc. v. Commissioner (T.C. Summary Opinion 2013-62), the IRS expert built his figure on a 2,080-hour work year (a standard 40-hour week, though the shareholder often worked 12-hour days with few days off); the Tax Court set $40/hour ($83,200). The court adjusted the hourly rate, not the hours — 2,080 was the IRS expert’s full-time-year basis, not an hours cap the court imposed.

Owners who work fewer than 2,080 hours annually may enter their actual hours. Compensation scales proportionally: an owner working 1,040 hours (half-time) would receive approximately half the full-time compensation figure, as each task’s hours allocation is reduced accordingly.

BLS OEWS wage estimates carry a May reference date and are published roughly a year later; the survey pools six semiannual panels (May and November) collected over three years. For reports covering tax years after the survey date, a data currency adjustment is applied to account for wage growth in the intervening period.

The adjustment uses the Bureau of Labor Statistics Employment Cost Index (ECI), specifically the index level for wages and salaries, private industry workers (BLS Series CIU2020000000000I, not seasonally adjusted). The ECI measures the change in the hourly labor cost to employers over time, using a fixed basket of labor so the result is free from the effects of workers moving between occupations and industries.

BLS publishes that series once a quarter. Each reading is the index level for the pay period that includes the 12th of March, June, September or December. The adjustment is the ratio between two of those readings: the index level at the date the wage is carried forward to, divided by the index level at the date the wage was surveyed. This is the method BLS sets out for aging survey wage data.

An OEWS release carries a May reference date, and this report sets compensation for a tax year, which begins on January 1. Each of those two dates falls between two quarterly readings, so each is placed between the readings on either side of it, one month at a time. May is two of the three months from the March reading to the June reading, so the index at May is the March reading plus two thirds of the step from March to June. January is one of the three months from the December reading to the March reading, so the index at January is the December reading plus one third of the step from December to March.

This tool implements all three valuation approaches described in the IRS Job Aid: Cost, Market, and Income. The Income Approach (Independent Investor Test) is available on CPA and tax-professional accounts at every plan level and considers the company’s financial performance via Fair Market Value data (see Section 05).

The following — five items from IRS Fact Sheet FS-2008-25’s nine-factor list that the tool does not directly compute, together with two related considerations outside that list — fall outside the tool’s scope and should be evaluated by the owner’s tax professional: