Economist · The scholar of scarcity — from Adam Smith's pin factory to the central-bank decision room, still asked to predict what no model fully captures.
From outside, the economist's skill looks like mathematics. Inside the profession, the mathematics is table stakes; the scarce skills are judgment about data — knowing what a number can and cannot support — and the discipline of stating, out loud, how uncertain a conclusion really is. The best-regarded economists are rarely the best mathematicians; they are the ones who ask questions the data can actually answer.
The craft divides into a research core — framing a question, finding credible variation, estimating and stress-testing an effect — and a communication shell: turning the result into two pages a minister, governor or CEO will read and act on. Careers stall on the shell as often as on the core, which is why the profession's private advice literature is mostly about writing and simplicity.
Estimating effects from messy real-world data, and knowing which of the profession's many statistical traps a given dataset is setting.
Giving usable advice when the model is known to be incomplete — the skill crises reveal, and the one seniority is actually paid for.
Building the smallest formal model that clarifies a question, and knowing when a model is decorating a paper rather than disciplining it.
Most working hours are spent in Stata, R or Python cleaning and reshaping data; empirical results are only as good as this unglamorous layer.
Research lives or dies by the paper; policy advice lives or dies by the two-page brief and the sixty-second answer to a decision-maker's question.
Knowing how the numbers are built, how the last three crises actually unfolded, and which lessons the models quietly assume away.
Major statistics land early — inflation, employment, GDP — and the first hours go to reading them against expectations, alongside overnight market moves and the working papers worth skimming.
The core production hours: cleaning data, running and re-running estimations, debugging model code, checking whether yesterday's result survives an alternative specification.
The brown-bag seminar is the profession's daily gym: a colleague or visitor presents work in progress and takes interruptions from the first slide, in academia and central banks alike.
Turning analysis into output — drafting the forecast note or paper section, briefing a chief economist or policy committee, refereeing, supervising research assistants, teaching on academic days.
The field moves through working papers long before journals; staying current is a standing evening tax, along with co-author calls across time zones.
The rhythm spikes around policy meetings, budget nights and market turmoil: forecast rounds before a rate decision or a crisis weekend can consume nights that the calendar had marked as rest.
Craft knowledge practitioners actually pass on — not motivation.
The modern field's first commandment: before estimating anything, name the source of variation that makes the comparison causal rather than coincidental — a lottery, a policy border, a sudden rule change. The "credibility revolution" made this the standard by which empirical work is admitted into the conversation at all.
Before trusting a series, learn who collected it, how the question was asked, and what changed when the methodology was revised — because the flaws in economic data are not random noise but systematic artifacts of collection. Griliches spent a career showing that the data constraint, not the econometrics, is usually what limits what economists can honestly claim.
Krugman's written rules for research — question the question, dare to be silly, simplify relentlessly — describe how his Nobel-winning trade models began as toy examples colleagues found embarrassingly simple. The craft is stripping a problem to the minimal model that still produces the surprising result, then stopping.
Friedman's methodological essay argued that unrealistic assumptions are not a defect: a model is a tool, and the only fair test is whether it predicts better than the alternatives. Working economists apply it daily as triage — challenge a model's forecasts, not its resemblance to reality.
The profession's oldest working reflex: decisions are made at the edge — the next unit, the next hour, the next euro — so sunk costs are ignored and averages distrusted. Marshall built the discipline's teaching around marginal analysis, and it remains the fastest way to spot a bad argument in a meeting.
The Bank of England's fan charts, introduced in its 1996 Inflation Report, print the forecast as widening bands of probability rather than a single line — institutionalizing the craft rule that an honest forecast states its own uncertainty. Senior forecasters teach juniors to say the band aloud before the central number.
The workhorse statistical package of applied economics since 1985, standard in empirical microeconomics and at institutions like the World Bank; entire replication archives of the field's top journals are written in it.
The open-source pair steadily displacing proprietary software: R for statistics and graphics, Python for data pipelines and machine learning, both now expected of new hires in central banks and tech-sector economist teams alike.
The open-source toolbox for solving and estimating DSGE models — the dynamic stochastic general equilibrium workhorses of academic macroeconomics — used for forecasting and scenario work at dozens of central banks.
The St. Louis Fed's free Federal Reserve Economic Data service, hosting hundreds of thousands of time series from around the world; its charts are the profession's lingua franca on social media and in seminar slides.
The market economist's cockpit since 1982: live prices, economic calendars, consensus forecasts and the chat network where financial economists' calls are made, tested and mocked in real time.
The field's canonical failure: a well-fitted regression on observational data, published with causal language, that collapses when a confounder surfaces. The replication and credibility movements exist because decades of influential results — on growth, aid, nutrition, policing — did not survive better identification.
Reporting a point forecast or a headline elasticity without its uncertainty band is the profession's most public failure mode — the 2008 crisis, foreseen by almost no mainstream model, prompted Queen Elizabeth II's famous question at the LSE and a decade of institutional soul-searching about overconfident precision.
Assumptions adopted for tractability harden into beliefs about the world; the economist starts defending the model against inconvenient data rather than revising it. Every generation relearns this expensively — the Phillips curve in the 1970s, efficient markets in 2008 — and the discipline's internal critics call it the field's characteristic occupational disease.
The person who starts the company — spotting the gap, bearing the risk and answering for payroll, from Assyrian caravan financiers to venture-backed founders.
AI-resistant 88 💹The dealmaker who prices companies and moves capital — a trade running from Medici Florence to today's pitch decks, paid for trust when billions change hands.
AI-resistant 42 🧾The keeper of the books: heir to a craft so old it invented writing itself, now negotiating with the software built to automate it.
AI-resistant 35 📣The professional who creates demand — from Pompeii's painted walls and P&G's 1931 brand-man memo to the auction-driven feeds of the digital era.
AI-resistant 38 🗺️Decides what a company should build next, and why — turning customer needs, business goals and engineering limits into one shared plan nobody else fully owns.
AI-resistant 50