BLS OEWS as a damages source: what occupational wage data can and cannot tell a forensic model
The Occupational Employment and Wage Statistics program is the backbone of imputed-mitigation analysis in most employment damages matters. Using it well means knowing where its limits sit.
The Bureau of Labor Statistics Occupational Employment and Wage Statistics (OEWS) program is the most cited labor-market data source in forensic economics. It produces occupation-specific wage estimates for the nation, each state, and most metropolitan and non-metropolitan areas. For imputed-mitigation analysis, OEWS is often the starting point. It is rarely the ending point.
What OEWS is
OEWS is a semi-annual survey of employer establishments, collecting wage and employment data for roughly 800 Standard Occupational Classification (SOC) occupations. The published figures are estimates of hourly and annual wages at the 10th, 25th, 50th, 75th, and 90th percentiles, along with employment counts. Geography is available at the national, state, and metropolitan statistical area (MSA) level, and for some substate non-metropolitan aggregations.
What OEWS can tell you
OEWS is representative, in the statistical sense: the survey design captures a broad cross-section of employers across industries. For a generic professional occupation in a large MSA, the OEWS median is usually a fair central-tendency estimate of what someone in that role earns.
For imputed mitigation, OEWS gives the economic model a defensible statement about what a plaintiff in a given occupation and geography could reasonably expect to earn. The report cites the most recent OEWS vintage, the specific SOC code, and the geography used.
What OEWS cannot tell you
OEWS has four well-known limits that matter in forensic work.
The median masks the distribution. For a plaintiff at the 90th percentile of a broad occupational category, the OEWS median understates the plaintiff’s earnings potential. The percentile series helps, but the top percentile (90th) is a censored estimate and the true 99th or 95th percentile earners in the occupation are not individually identified.
Equity compensation is absent. OEWS captures wage and salary income. Stock-based compensation, deferred cash, and long-term incentive payouts are not in the wage data. For executive and senior-professional plaintiffs, OEWS will materially understate the total compensation picture.
Industry variation is not disaggregated within most SOC codes. A senior software engineer in finance, a senior software engineer in healthcare IT, and a senior software engineer in a video game studio have very different pay norms. OEWS reports a single aggregate for “Software Developers” (15-1252). For plaintiffs in particular industry niches, OEWS must be supplemented with industry-specific sources.
Firm-size effects are not in the data. Compensation at the 90th percentile Fortune 500 firm differs from compensation at the 90th percentile mid-market firm in the same occupation and MSA. OEWS does not disaggregate.
Using OEWS well
The forensic-economic report that uses OEWS well does three things. It cites the specific OEWS vintage, SOC code, and geography in a footnote. It states the percentile used, and why. And it pairs OEWS with at least one supplementary source appropriate to the plaintiff’s profile: an industry compensation survey, a firm-size adjustment, or a proxy based on public-company filings.
OEWS is a foundation. On top of a foundation, you build.