$299,110top of the range in District of Columbia · middle $124,720 / yr
AI augments this role
Economists in the United States earn a median of $124,720 a year. Pay starts near $67,360. Pay reaches $299,110 at the top of the range in Washington D.C., the best-paying location for this work among those with at least 500 people in the job.
Source: U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025 (Economists, SOC 19-3011). Last checked 9 September 2026.
Entry level
$67,360
Top of the range · District of Columbia
$299,110
Education
Master's or Doctoral degree in Economics
Wages — U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025 (Economists). Top of the range is the highest state-level figure among states with at least 500 people in the job. AI-impact rating is PayCrunch's editorial assessment. Updated September 2026.
🆕 New & Trending AI Tools for EconomistReviewed September 2026
We track new AI-tool launches every week and refresh this list — here’s what’s gaining traction for Economist work right now.
NumericNEWPaid / see site
AI-driven month-end close, reconciliation, and reporting.
How an Economist uses it: automate reconciliations and close the books faster
HebbiaNEWEnterprise / see site
AI that reads and analyzes large financial documents and filings.
How an Economist uses it: pull answers out of contracts, filings, and reports in minutes
NotebookLMNEWFree / $7.99 mo
Google tool that answers questions grounded only in the documents you give it — with citations.
How an Economist uses it: load your own manuals, policies, or PDFs and ask questions that stay accurate to the source
MindBridgeEnterprise / see site
AI that scans transactions for anomalies, errors, and fraud risk.
How an Economist uses it: flag risky or unusual entries across the whole ledger, not just a sample
Vic.aiEnterprise / see site
Autonomous accounts-payable and invoice processing.
How an Economist uses it: let AI code and process invoices with minimal manual entry
RampFree core / paid
Finance platform with AI that automates expenses and spend controls.
How an Economist uses it: auto-categorize spend and catch policy issues in real time
Power BI Copilot$10+ mo
Microsoft analytics with AI that builds dashboards and explains trends.
How an Economist uses it: ask questions of financial data and get charts and forecasts back
ChatGPTFree / $20 mo
The most-used AI assistant — writing, analysis, research, and images from a plain-language chat.
How an Economist uses it: draft emails and documents, summarize long files, and get instant answers to on-the-job questions
ClaudeFree / $20 mo
AI assistant known for careful writing, long-document analysis, and coding.
How an Economist uses it: analyze big reports or spreadsheets and turn messy notes into clean, finished writing
The chart is due before the meeting, and someone in that meeting will make a decision with it. You check the series again, you check the footnote that explains a break in the data, and you rewrite the sentence that sounded certain and was only almost certain. An economist in this line of work produces analysis other people use: a forecast, an evaluation, a pricing recommendation, a memo on what a policy is likely to do. The day is not organized around a course catalog. It is organized around a decision, a client, or a public brief that has to be defensible when a skeptic reads the appendix.
The work is the analysis, not the lecture hall
Government economists staff agencies, central banks, legislatures, and budget shops. They frame what a proposal costs, who is affected, and what the evidence can and cannot support. Consulting economists do related work for hire: litigation support, antitrust, health, energy, labor, or a regulatory filing a client must submit. Private firms hire economists to study demand, set prices, evaluate a program, or build the internal view of the economy their executives will quote. The tools overlap. The audience changes the tone. A memo for a deputy secretary and a deck for a product lead can use the same regression and should not sound like the same document.
A normal week mixes data work with conversation. You pull a series, you clean it, you run the specification you pre-committed to rather than the one that flatters the result, and you write what the result means in words a non-specialist can use. You sit with lawyers, engineers, program managers, or traders, depending on the shop, and you translate. You also push back when someone wants a number the design cannot produce. That pushback is the job. An economist who only says yes becomes a decorator for a decision that was already made. An economist who only says no never ships anything a manager can use on Thursday.
Research seats inside a bank, a think tank, an international organization, or a lab look quieter and are not actually quiet. You own a longer project, you present it to peers who will try to break it, and you revise. The output is a paper, a working report, or a model the institution stands behind. Deadlines still exist. They are just spaced differently from a weekly briefing. If you want a classroom, a tenure file, and office hours, you want a faculty appointment. This occupation is the applied or research seat: the analysis is the product, and students are not the organizing principle of the week.
A master's for many seats, a doctorate for many research ones
A master's degree is common preparation for many working economist seats, especially in government, consulting, and firms that need strong empirical skills without a dissertation. The degree shows graduate training in theory and methods and, at its best, a finished piece of applied work. People prepare by choosing a program that makes them write code, handle real data, and explain results, then by internships or analyst roles that prove they can meet a deadline that is not a syllabus date. Grades matter less, in hiring, than whether you can reproduce your own tables.
A doctorate is common in research roles: research departments, many think-tank posts, and jobs that expect you to publish or to lead a technical agenda. The Ph.D. shows you completed an original project and survived expert criticism. It is a degree, not a professional licence. No board cards you as an economist the way a bar cards a lawyer. Employers hire the training and the record. Some postings will take a master's with unusual experience in place of a doctorate, and some will not. Read the posting literally. If it says Ph.D. required for the research track and master's accepted for the associate track, believe the split and apply to the track you can actually fill.
Degree, not a card
Many seats hire from a master's. Research seats often expect a doctorate. Neither credential is a national licence to practice economics.
Getting hired in an agency, a firm, or a consultancy
Applications are evidence. Include a writing sample that is yours, short enough to be read, and clear about what you did versus what a coauthor did. Include code or a technical appendix if the job is empirical and you are allowed to share it. Be ready to walk through a project from the raw problem to the caveat. Interviewers can smell a memorized result. They trust a candidate who can say what broke, what you changed, and what you still do not know. For government roles, follow the posting's process exactly, including any public-service format it demands. For consulting, expect a case-style conversation plus a look at your writing. For a company, expect a discussion of a product or a market, and a demand that you tie the answer to a decision.
The first year is mostly other people's projects. You learn where the data live, which caveats the institution always wants, and how a senior economist likes a chart. You will redo a table because a merge was wrong. That is a better lesson than a perfect first draft that nobody checks. Ask to draft the recommendation section once your descriptive work is trusted. The people who advance are the ones who can be left with a messy dataset and a decision-maker's actual dilemma, and who come back with a result that matches the code.
Communication is part of the craft, and it is easier to fake in an interview than on the job. A clean chart with a dishonest axis will be noticed by the one person in the room who knows the series. A careful caveat, written in ordinary words, will be remembered the day a result gets quoted out of context. Practice saying the finding in three sentences: what you measured, what you found, and what would change your mind. Seniors promote economists who can do that without turning the caveat into a fog. They also notice who updates a number when the data revision lands, instead of defending the old slide because it already went to the client.
Analyst, economist, then the person who sets the agenda
Titles drift by employer. Analyst and research assistant seats support. Economist and senior economist seats own a topic. A principal, a partner, or a research director sets scope, reviews other people's work, and faces the client or the political principal when the finding is awkward. Government ladders add management: a unit chief who still understands the method. Consulting ladders add business development, which is a real change in the job, not a hobby on top of research. Private firms sometimes fold the economist into a data-science or strategy title. Read the tasks. If you are forecasting, evaluating, and explaining tradeoffs, you are in this occupation even when the badge says something trendier.
Moves between sectors are common and should be deliberate. A government economist who joins a consultancy trades a public mission for client pace and, often, a different pay structure. A consulting economist who joins a firm trades variety for depth in one business. A research economist who wants a classroom later is changing occupations, not just employers, and should expect a faculty search rather than a lateral title match. Keep a portfolio of public or shareable work so the next sector can see your judgment. Confidential client work still counts in an interview if you can discuss the method without disclosing the client. Practice that distinction before you need it.
May 2025 pay for the Economists title
The May 2025 Occupational Employment and Wage Statistics title Economists matches this occupation. Entry pay is $67,360. The national median is $124,720. High end of what the Bureau published is $299,110 in the District of Columbia, in the locations with enough economists for the Bureau to report a high end. The District's median is a different statistic: $167,590. If you remember only one caution, remember that $299,110 is the top of the range and $167,590 is the middle of District wages. They answer different comparisons.
The District leads state medians at $167,590, which is $42,870 above the national median. Maryland's median is $144,680. Virginia's is $137,590. New York's is $136,660. Texas is $121,420. Florida shows the lowest state median published here, and the gap from the District's median down to that low end is $65,160. The corridor of the District, Maryland, and Virginia is a real concentration of this work, and the medians show it without requiring you to treat the range top as a typical paycheck. New York's $136,660 is strong and still well under the District median. Texas at $121,420 sits near the national median of $124,720.
Entry to the national median is $57,360. That gap is the national picture of moving from a junior seat toward the middle of the occupation. The national median to the District's range high end is $174,390, a much longer stretch, and it describes the far published edge rather than a standard promotion. Use $67,360 to test an offer for an economist who still works under close review. Use $124,720 to test a full economist at the national middle. Use $167,590 when the job is in the District and you mean the median. Use Maryland, Virginia, New York, or Texas medians when the offer is in those places. Use $299,110 only with the words high end of the range attached, and only for a claim about the top of published pay.
When the offer shows up, match it to the seat
Convert bonus-heavy offers into a base you can verify and a bonus you treat as uncertain. Consulting and some firms pay a base plus a bonus that depends on the year. Compare the base with $67,360 or $124,720 before you mentally spend the bonus. If the role is a research doctorate seat with a publication expectation and little supervision, an entry-level base is the wrong anchor even if the title is modest. The $57,360 from entry to the national median is the step that matches independent ownership of a workstream. Say so with the duties, not with a speech about what economists "should" earn.
Location is a separate sentence. The District median is $42,870 above the national median. That is the geographic gap on the median statistic, not a personal compliment and not the $299,110 range figure. Maryland at $144,680 and Virginia at $137,590 are the numbers to use if the office is in those states and you commute into a regional labor market that is not the District itself. New York at $136,660 and Texas at $121,420 give you local medians so you do not import a District comparison by accident. Florida's low end, $65,160 below the District median, is the reminder that state middles are not interchangeable. Housing costs are outside these figures. Do not invent a cost-of-living add-on and call it a Bureau number.
A raise inside an institution works best when your scope changed. You moved from supporting a senior economist to owning a brief. You started reviewing other analysts. You became the person a principal calls when a filing is due. The national median is the checkpoint for that ownership. A principal or partner conversation can look beyond it, and the state median is often the cleaner local checkpoint than the range top. The $174,390 from the median to the District range high end belongs to a discussion of the far end of published wages, typically senior research leadership in a high-paying market. Bring $299,110 into that discussion only if the role matches the claim. Bring $124,720 into the ordinary one.
Keep the occupation clear while you negotiate. You were hired to analyze, to write, and to withstand a skeptic. A master's opened many of these seats. A doctorate opened many of the research ones. Neither one licensed you. Faculty life, with its courses and its tenure file, is a different occupation. The May 2025 Economists figures are the ones that match this work: entry $67,360, national median $124,720, District median $167,590, and a high end of the range at $299,110 in the District. Quote the one that matches the offer in front of you, and leave the others in the notebook so they cannot be swapped mid-sentence by a hurried recruiter.
The top of Economist pay — and how to get there with AI
$299,110what Economist pay reaches in District of Columbia
Highest state-level top-of-range annual wage for Economists, among states with at least 500 people in the job. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025.
And the role it leads to — Financial Managers — reaches $370,780 in New York.
$67,360entry$124,720middle$299,110top end
Economists paid at the top of this range are the ones whose models rerun themselves whenever new data lands, so their time goes into the impact statement and the judgment inside it rather than into rebuilding a table.
Look honestly at where a week goes. Monitoring market and environmental trends, pulling series, re-estimating, rebuilding exhibits, then writing the technical document, the impact statement or the grant proposal around them. Estimation is the interesting hour; the rest is assembly. Once your system for collecting, analyzing and interpreting data is genuinely a system, with scripted pulls, versioned model code and exhibits regenerated on command, a forecast update stops being a week of work. What remains is what agencies and clients are actually short of: somebody who can say what the number means for a policy and then defend it.
Your playbook, by where you are now
Just startingMake the analysis rerun on command
Script every data pull and transformation, so a refreshed series regenerates the whole exhibit set with no hand editing.
Keep model code under version control with its estimation output stored alongside, so any published figure can be traced backwards.
Learn one econometric package thoroughly, whether IBM SPSS Statistics or Aptech Systems GAUSS, rather than three of them badly.
Write the methods section while you build the model, not afterwards, while you still remember which assumptions you made.
Let a model draft the descriptive paragraphs around your tables, then check every magnitude and sign against the actual output.
What proves it: A study another analyst can reproduce end to end from your files.
Realistic span: your first three years
A few years inOwn a recurring product
Take charge of one regular publication, a forecast, an outlook or a market monitor, and industrialise it until an update takes an afternoon.
Build the exhibit library once so the chart pack rebuilds itself each cycle instead of being reassembled in Microsoft Excel.
Add spatial work in ESRI ArcGIS wherever resource and environmental questions need it, since maps carry arguments that tables cannot.
Write research proposals and grant applications from a reusable core, so each new one is an edit rather than a blank page.
Deliver the presentation yourself whenever the chance appears, because being the voice on a forecast is what makes it yours.
What proves it: A recurring published product you own and can refresh in a single day.
Realistic span: years four to eight
ExperiencedSell judgment, not tables
Write the impact statements and long-horizon assessments, rehabilitation costs and resource exhaustibility, where modelling and policy meet.
Do the integrated mathematical modelling nobody else in the shop will attempt, and document its assumptions so it survives you.
Teach a course or run internal training, which is how senior people in a research shop build reputation cheaply.
Move toward roles where your analysis drives budgets and capital decisions, since financial management is the common step up from an economist post.
Watch where policy demand concentrates, because the District of Columbia pays this occupation the most and federal work explains it.
What proves it: An impact assessment or policy recommendation adopted by the body that commissioned it.
Realistic span: nine years and beyond
The next 90 days
Choose the analysis you rerun most often and rebuild it once, within ninety days, as a proper pipeline: raw data in, cleaning and estimation scripted, every table and figure produced by code, and the document's numeric claims read from the output rather than typed in. Do it properly even though the first pass costs more than doing it by hand would. The next update takes an afternoon and the one after that takes an hour. Then spend the difference where an economist is genuinely scarce, reading results against the policy question and writing the paragraph that says what ought to change.
Wage figures: BLS OEWS, May 2025. The playbook is PayCrunch editorial guidance, not a guarantee of pay or placement.
Every figure is the national median from the U.S. Bureau of Labor Statistics (OEWS) shown on that role’s own page.
Never used AI before? Start here (2 minutes).
Start by turning natural language into working econometric code. Open Claude or ChatGPT (with data-analysis/code tools) and use it to write and debug the Python, R, or Stata behind your models — regressions, fixed effects, instrumental variables — while you own the specification. Pull real series from FRED and let AI script the cleaning and estimation you'd otherwise hand-code.
For literature, use Elicit or Consensus to find and summarize papers (then read the originals), and Perplexity for cited factual research you verify. Keep confidential data in approved environments. AI is the research assistant who codes and drafts; you are the economist who owns the identification and the interpretation.
The one rule, forever: In research, a fabricated number or citation is career-ending — AI hallucinates both. Never trust an AI-provided statistic, paper, or dataset without verifying it at the primary source, and never let AI invent data. Keep confidential, embargoed, or proprietary datasets out of consumer tools and inside approved environments. Every analysis must be reproducible and every causal claim must rest on an identification strategy you can defend; AI writes the code, but the economic reasoning is yours.
The plays — exact steps, exact prompts
Do these in order. Each one is copy-paste ready. You do not need to know anything about AI going in.
1
Turn econometric ideas into code at the speed of thought
Why this pays: The economists who command top pay are the ones who can actually implement sophisticated analysis, not just describe it. AI collapses the coding barrier so you can run more, and more advanced, models — the analytical productivity that defines a senior tech or consulting economist.
ClaudePythonStata
1
Describe your model to Claude and have it write the Python (statsmodels/linearmodels) or Stata code, including the robustness checks — then read every line and confirm the specification is what you intended.
2
Use AI to translate between languages and methods so you're never blocked by tooling.
Copy-paste this prompt
You are an econometrics coding assistant. I want to estimate [a difference-in-differences model with staggered treatment timing] on panel data with columns [unit, time, treatment, outcome, controls]. Write commented Python using a modern estimator that handles staggered adoption (e.g. Callaway-Sant'Anna), include the parallel-trends check and event-study plot, and explain each modeling choice and its assumptions so I can defend them.
You own the specification and its assumptions — verify the estimator fits your identification, and never let AI choose the design for you.
3
Keep a reproducible script for every analysis. Reproducibility is what separates credible economics from a black box.
What you'll haveAdvanced models implemented fast and reproducibly — the analytical output that earns senior economist pay.
2
Master modern causal inference with AI-assisted ML
Why this pays: Causal inference on big data is the single most valuable economist skill in industry — it's what tech companies hire economists for. Using AI to become fluent in the modern causal-ML toolkit puts you in the highest-paid segment of the profession.
EconMLDoWhyClaude
1
Learn the modern causal-ML stack — EconML (heterogeneous treatment effects) and DoWhy (identification and refutation) — with Claude explaining the methods and writing the starter code as you go.
2
Have AI help you apply the right estimator and, crucially, test whether the causal claim holds up.
Copy-paste this prompt
You are a causal inference expert. I want to estimate the heterogeneous effect of [a pricing change] on [customer retention] using observational data with confounders [list]. Recommend an appropriate EconML estimator (e.g. Double ML / causal forest), write the code, and then design the refutation tests (placebo, random common cause, subset) using DoWhy to check robustness. Explain what would invalidate the causal claim.
Causal validity rests on your identification assumptions, not the library — run the refutation tests and be honest about what could break the claim.
3
Build a portfolio of clean causal analyses. Demonstrated causal-ML skill is the ticket into the best-paid economist roles.
What you'll haveFluency in the causal-ML toolkit industry pays most for — the skill behind the tech-economist salary tier.
3
Design and analyze experiments like a tech economist
Why this pays: Large companies run thousands of experiments and need economists who can design them and read the results correctly. Being the person who owns experimentation — power, effects, interference — is a high-leverage, high-comp role AI helps you grow into fast.
PythonClaudeR
1
Use Claude to build and check your experiment design — power analysis, minimum detectable effect, variance reduction (CUPED) — and to write the analysis code in Python or R.
2
Pressure-test the design and the pitfalls before the experiment runs.
Copy-paste this prompt
Act as an experimentation economist at a large tech company. I'm designing an A/B test to measure the effect of [a feature change] on [a conversion metric]. Help me: compute the sample size for a minimum detectable effect of [X%] at 80% power, identify threats (network interference, novelty effects, multiple testing), recommend a variance-reduction approach, and specify the analysis so the estimate is unbiased. Explain the trade-offs.
Design decisions and threat assessment are yours to own; AI can compute and suggest but can't judge whether the estimand answers the business question.
3
Present results with honest uncertainty. Economists who quantify what we can and can't conclude are trusted with the biggest decisions.
What you'll haveRigorous experiment design and analysis — the experimentation ownership that defines the high-paid industry economist.
4
Compress the literature review and horizon scan
Why this pays: Whether in policy, consulting, or industry, being current and comprehensive fast is a competitive edge. AI turns weeks of literature work into days, so you produce deeper, better-grounded analysis — the quality that earns senior and advisory roles.
ElicitConsensusNotebookLM
1
Use Elicit or Consensus to map the literature on a question — find the key papers, extract methods and findings into a table — then read the originals before you cite anything.
2
Load the core papers into NotebookLM and interrogate them for the debate and the gaps.
Copy-paste this prompt
I'm reviewing the literature on [the employment effects of minimum wage increases]. From these papers [load as sources], summarize the main findings, the methodological approaches and how they differ, where the evidence conflicts and why, the identification strategies used, and the open questions a new study could address. Cite which paper each claim comes from so I can verify.
Verify every citation and finding against the actual paper — AI misattributes and invents references. Never cite what you haven't read.
3
Synthesize it into your own framing. The value is your judgment about what the evidence means, not the summary itself.
What you'll haveComprehensive, current, verified literature grounding fast — the depth that underpins senior and advisory work.
5
Write reports and briefs that decision-makers act on
Why this pays: An economist's influence — and pay — depends on translating analysis into decisions executives, regulators, or clients actually make. AI helps you turn dense results into clear, persuasive writing for non-economists, multiplying the impact of every analysis.
ClaudeChatGPTMicrosoft Copilot
1
Draft the plain-language version of your findings in Claude, keeping the technical rigor in an appendix, so a busy decision-maker gets the 'so what' immediately.
2
Translate a technical result into an executive brief without losing the caveats.
Copy-paste this prompt
You are an economics communicator. Turn this technical finding into a one-page brief for a [non-economist executive / policymaker]: [state your result, method, and key caveats in plain terms]. Lead with the decision-relevant takeaway, explain the magnitude in intuitive terms, state the main assumptions and uncertainty honestly, and end with the options and trade-offs. No jargon, no overclaiming.
Never let the simplification overstate certainty — preserve the caveats and the limits of what the analysis shows.
3
Tailor the same analysis to each audience. The economist whose work drives decisions is the one who gets the advisory seat and its pay.
What you'll haveAnalysis that decision-makers actually act on — the influence and reach that lift an economist to the top of the band.
Your 12-month sequence to the top of the range
How the plays above stack into a path from median pay toward the $299,110 tier.
Month 1
Use AI to write and debug your econometric code (Python/R/Stata) while you own every specification; keep reproducible scripts.
Months 2-3
Build fluency in modern causal-ML (EconML, DoWhy), including the refutation tests that check whether a causal claim holds.
Months 3-6
Learn to design and analyze experiments like a tech economist — power, variance reduction, interference — and build a portfolio.
Months 6-12
Speed up literature reviews with verified AI research, and sharpen turning analysis into decision-ready briefs.
Year 2
Combine causal-ML and experimentation skill into industry or consulting roles — the tech-economist and advisory tier that pays $299k.
Gear for this job
As an Amazon Associate, PayCrunch earns from qualifying purchases. Links to books and tools are for the job on this page; we only recommend what we’d use in the work.
Same live O’Reilly 3rd already on data-scientist / python-developer / market-research-analyst / quantitative-analyst. This leftover page’s first play is Turn econometric ideas into code at the speed of thought and names Python (statsmodels/linearmodels); Month 1 is Use AI to write and debug your econometric code (Python/R/Stata). Not leftover 94 CFP and not CFA Level I as the lead (that is financial-analyst / credit-analyst). Confirm 109810403X. Live page HTTP 200, no PC_GEAR / amazon.com/dp / tag=paycrunch-20 at 2026-09-17 3:31 PM PT.
Next steps for an Economist
Some links below are affiliate or partner links. PayCrunch may earn a commission if you enroll or subscribe through them, at no extra cost to you. Wage figures on this page still come from the Bureau of Labor Statistics, not from these programs.
Economist work is specific enough that a stamped 'check out these courses' block would be noise. BLS files this work as Economists (SOC 19-3011). O*NET Job Zone 5 is typical: graduate or professional school, so the honest next credential is a graduate-level or professional certificate — not a random catalog dump.
The occupation's listed knowledge area is Economics and Accounting, which is what the course searches below actually query.
Economists in this dataset list C among the tools in use, so a program that names that stack is a better fit than a survey course.
Coursera search for economics and accounting — a graduate-level or professional certificate that lines up with science, not a generic professional-development aisle.
FlexJobs screens remote, hybrid, freelance, and flexible listings so you are not wading through unverified ads. This is a job-board search for Economist work, not a claim that they list a counted SOC 19-3011 inventory.
Write an Economist resume, or one aimed at Financial Managers, instead of a blank template. Resume Now is a resume builder; we are not claiming a counted template set for this SOC.
An Economist resume that names the actual tasks on this page, or the step-up title Financial Managers, beats a blank template when you apply.
What Economists earn by state
These are the Bureau of Labor Statistics’ own figures for Economists, state by state — not a cost-of-living adjustment applied to the national number. Only states employing at least 500 people in the occupation are shown, because a state median drawn from a handful of workers is noise rather than a signal.
District of Columbia
$167,590
highest of them · +34% vs the national median
Florida
$102,430
lowest of the 10 states and D.C. that qualify · -18% vs the national median
The same job pays $65,160 more a year at the median in District of Columbia than in Florida — 64% higher. That gap is what the Bureau measured, before any question of what it costs to live in either place. District of Columbia also carries the top of this job’s range, $299,110 — the figure quoted at the head of this page.
Source: U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025, SOC 19-3011. 10 states and D.C. clear the 500-employee reporting floor for this occupation; those below it are left out rather than shown with a wide error band.
Free data. Use any of it.
PayCrunch publishes verified, BLS-sourced salary + AI-playbook data on 1,000+ professions — free, no signup.
No — it shifts where the value sits. AI can run a regression and draft a summary, but choosing a credible identification strategy, judging whether an estimate is causal, interpreting what a result means for a real decision, and standing behind it are human accountabilities. Economics is about credible inference under uncertainty, not just computation. The economists who use AI to run more and better analysis — especially causal inference and experimentation — will be more valuable, and those are exactly the highest-paid roles.
Can I trust AI for data, statistics, and citations?
Never without verification. AI hallucinates statistics, invents plausible-looking papers, and misattributes findings — and in research, a fabricated number or citation ends careers. Use AI to find and draft, then confirm every figure, dataset, and reference at the primary source. Treat AI output as a lead to check, not a fact to cite.
Is it safe to put my data into ChatGPT or Claude?
Only if the data is public or fully de-identified. Confidential, embargoed, or proprietary datasets belong in approved secure environments, not consumer tools. You can safely use AI to write code against public data (like FRED) or to reason about anonymized structures, but keep restricted data where your institution's policies require.
How does AI actually raise an economist's pay?
By moving you toward the highest-paid work. AI collapses the coding barrier so you can run advanced causal-ML and experimentation — precisely the skills tech companies and top consultancies hire economists for, at salaries well above academia or government. It also speeds literature work and makes your writing decision-ready. More rigorous analysis, communicated to decision-makers, is what earns the tech-economist and advisory tier.
Do I still need the PhD if AI can do the technical work?
For the top roles, usually yes — the PhD signals you can design credible inference and defend it, which is exactly the judgment AI can't replace and employers hire on. But AI dramatically raises what you can do at any level and makes the causal-ML and experimentation skills learnable faster. Use it to deepen the identification and interpretation skills the degree is really about, not to skip them.
Methodology & sources
Salary (median, 10th, top of the range) — U.S. Bureau of Labor Statistics, OEWS.
By state — the Bureau of Labor Statistics’ own state medians, limited to states employing at least 500 people in the occupation. No cost-of-living arithmetic is applied to a wage anywhere on this page.
The plays — PayCrunch's own step-by-step guidance using publicly available AI tools. Tool names/URLs are real and current as of August 2026; prompts written to work as-is. Verify any professional output before relying on it.