EXPOSED
The daily work of a median economist — cleaning datasets, running regressions and time-series models, writing literature reviews, producing forecast memos and chart-heavy briefs — sits squarely in the zone where LLMs plus statistical tooling already produce usable first drafts. What resists is causal identification judgment: choosing the instrument, defending the counterfactual, deciding which model failure matters, and putting your name on a forecast or damages estimate that a court, a central bank committee, or a CFO will act on. There is no license, so the moat is reputation and accountability rather than regulation, and it is thinner for the many economists doing routine industry analysis than for those testifying or advising policy.
Nearly all of this fall was the 2020 shock. It has been climbing back since.
Median pay $105,020 → $124,720 -5.0% in real terms
This line is counted by the Bureau of Labor Statistics — the one figure on this page that isn't a judgement of ours. Headcount moves on demand, offshoring, demographics and the business cycle, and automation is one term among several, often not the loudest.
So a falling line is not evidence that AI did it, and a rising one is not evidence that it won't. Both happen in this register: some occupations resist automation and shrink anyway, others are highly automatable and keep growing. The marked year is 2020.
BLS projection, 2024–2034
+1.2% 17,600 → 17,800 on the projections basis
Exposed, but growing
AI can already do a lot of these tasks, and the BLS still expects +1.2% more of these jobs by 2034. Demand for the output is growing faster than the work is being automated away — the mechanism BLS gives for software developers, and the combination people most often misread as an error.
Different clocks. The score is what current AI could do to this work today. The projection is how many of these jobs will exist in 2034. Everything between the two — how fast employers actually adopt, whether demand grows in the meantime — is why they can point opposite ways without either being wrong.
~900 openings a year on average, including replacing people who leave.
EconomistTax EconomistEconometricianHome EconomistPolicy AdvisorPolicy OfficerLabor EconomistPrice EconomistTrade EconomistEconomic AdvisorEconomic AnalystEnergy EconomistFiscal EconomistForecast AnalystForest EconomistHealth EconomistResearch AnalystSocial EconomistProject EconomistBusiness EconomistEconomic DeveloperForensic EconomistResearch EconomistResource Economist
Holding it up: judgment & accountability . Weakest point: embodiment .
Mixed — a routine tier and a judgment tier Pulling BLS/Census/Compustat series, coding a fixed-effects or VAR specification in Stata/R, drafting the literature review and the CPI-outlook memo are now first-draftable by machine, and an 8 rather than a 4 reflects that the identification step — arguing why your instrument is exogenous, why the pre-trends hold, why the 2020 structural break invalidates the pre-pandemic elasticity — still has to be reasoned by a person who will be cross-examined on it.
Fully desk- and screen-based The job is a laptop, a data extract, and a seminar room; the only physical duty is standing at a whiteboard or a hearing podium, which is why this is 1 and not 0.
No licence, no signature requirement There is no economist license, no exam, no board that can strike you off — a 3 rather than 0 only because expert-witness work runs through Daubert admissibility and federal disclosure rules that attach your name and CV to the damages calculation.
Meaningful discretion You choose the counterfactual, set the discount rate and the horizon, decide whether to strip an outlier quarter, and defend a number that becomes a rate vote, a merger remedy, or a nine-figure damages award — high stakes with genuine ambiguity, held below 14 because the model, the data vintage, and the peer-review or committee process absorb much of the blame when the forecast misses.
The verdict above describes this occupation as a whole. Almost nobody does the typical version of a job — tick what's actually in your week and see how your own mix sits.
Your task mix speaks to task resistance (8/20 here) — how much of the day's work current AI already does. That is the dimension the boxes above are about.
It cannot move the other three. Liability shield (3/20) is whether the law requires a licensed human to sign. Trust premium (10/20) is whether buyers specifically pay for a person. Judgment and accountability (13/20) is whether the role exists to own consequential calls. Those are facts about the occupation's standing, not about which tasks are in your week — a paralegal who does only trial exhibits still holds no licence. Together they are 26 of this occupation's 35 points (74%).
Embodiment (1/20) is also a property of the work rather than the worker, but we don't tag individual tasks as physical or not, so the picker can't tell you anything about it. That's a limit of this tool, not a claim.
Did we get the list right? Tell us what's missing — the tasks are written from the outside, and you're reading this from the inside.
Actuaries EXPOSED
Financial Managers EXPOSED
The moves above are yours to make. This is the other half: what would have to change in the world for the occupation itself to score higher. None of it is in any one person's gift, but it is where the floor actually comes from. Scores here are not a one-way ratchet. Only two of the five dimensions — task resistance and embodiment — track what machines can do. The other three track law, what buyers will pay for, and who is answerable, and those move in both directions, often in response to the same pressure AI creates. If every lever below landed, this occupation would score around 48/100, still EXPOSED.
Expert-witness economists already own damages estimates under Daubert; if courts tighten disclosure of AI use in expert reports (e.g. the proposed FRE 707 amendment on machine-generated evidence, published for comment 2024-25), the named economist becomes the accountable signer for model choice, instrument selection and counterfactual defense, and AI-only analysis becomes inadmissible without a human author.
Genuine two-tier structure: if data cleaning, regression running, literature review and chart-brief production are absorbed by tooling, the residual role is causal identification, model-failure triage, and defending a specification under cross-examination or before an FOMC-style committee — a task mix current models handle poorly because it requires committing to an untestable counterfactual.
Credentialing tied to signed work rather than a state license: if bank regulators extend model-risk-management rules (SR 11-7 / OCC 2011-12) to require a named qualified economist to attest to macro scenario and CECL model assumptions, or if actuarial-style attestation is imported for damages models, an identified human becomes a required signatory.
Buyers paying specifically for an attributable human name: central bank appointments, testifying-expert retention at hourly rates, and Article-level publication all price the named author's reputation. This premium is real but concentrated; if litigation-support and policy-advisory demand grows while routine industry forecasting commoditizes, the surviving segment carries it — it does not spread to the median industry analyst.
The limit. No realistic licensure route exists for economists as a whole; the CFA/actuarial analogy does not transfer. Most gains accrue only to the testifying/policy-advisory minority — for the many economists doing routine corporate forecasting and dashboard-adjacent analysis, none of these levers apply and the score stays near where it is.
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 4,570 | $156,750 +26% |
| New York-Newark-Jersey City, NY-NJ | 850 | $176,090 +41% |
| Boston-Cambridge-Newton, MA-NH | 820 | $108,140 -13% |
| Sacramento-Roseville-Folsom, CA | 250 | $100,410 -19% |
| Atlanta-Sandy Springs-Roswell, GA | 220 | $122,570 -2% |
| Philadelphia-Camden-Wilmington, PA-NJ-DE-MD | 220 | $139,530 +12% |
| Dallas-Fort Worth-Arlington, TX | 210 | $122,170 -2% |
| Los Angeles-Long Beach-Anaheim, CA | 200 | $151,510 +21% |
| New York-Newark-Jersey City, NY-NJ | 850 | $176,090 +41% |
| Washington-Arlington-Alexandria, DC-VA-MD-WV | 4,570 | $156,750 +26% |
| Baltimore-Columbia-Towson, MD | 110 | $152,730 +22% |
We have no reported case of a named organisation automating this occupation. Not one deployment, not one announcement.
That is worth saying out loud next to a score of 35. The verdict above is about what the work exposes — what current AI could do to these tasks. It is not a claim that anyone has done it. For this occupation those two things have come apart completely: the capability argument is on this page, and the evidence column is empty.
Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.
Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.