qje-topic-selection¶
Scopes and pressure-tests a research question for QJE — fit, the "big idea" bar, and breadth of the takeaway for QJE's general-interest readership.
rq-formulation
Topic Selection & Fit (qje-topic-selection)¶
When to trigger¶
- You have data or a result but the "so what?" is unclear
- You are unsure whether the idea is top-5 material or a field-journal paper
- The question is narrow, technical, or interesting only to specialists
- You can describe what you did but not the broad lesson it teaches
The QJE bar: a big question with a broad answer¶
QJE is a general-interest journal covering all of economics, published by OUP and edited at Harvard since 1886 (the oldest English-language economics journal). Its comparative advantage is the compelling question + credible empirics + a conceptual takeaway with broad implications — the journal that ran Akerlof's "Market for 'Lemons'" (QJE 1970) and Mankiw, Romer & Weil's "Contribution to the Empirics of Economic Growth" (QJE 1992) rewards papers that change how a wide audience thinks, not just precise estimates. Editorial reality reinforces this: five Harvard-based Editors desk-screen in ~2 weeks, so the big idea must be visible immediately. Force the idea through three filters:
- The question filter — Would a smart economist outside your subfield want to know the answer? State the question in one sentence a labor economist, a macroeconomist, and a development economist would all find interesting.
- The answer filter — Is the answer (not the method) surprising, important, or settling of a long-standing debate? A precise estimate of something nobody disputes is a field-journal paper.
- The breadth filter — Does the result generalize beyond the specific setting? QJE rewards a clean local result that teaches a general lesson (e.g., Chetty–Hendren–Kline–Saez on the geography of intergenerational mobility, QJE 2014).
Scope fit decision table¶
| Your paper is... | QJE fit | Better home if not |
|---|---|---|
| Big empirical-micro question + clean natural experiment | Strong | — |
| Novel large-scale data answering an old question | Strong | — |
| New econometric estimator / asymptotic theory | Weak | Econometrica, JoE |
| Structural IO / quantitative-macro led by a calibrated model | Possible | JPE, Econometrica, AEJ:Macro |
| Field-specific incremental estimate | Weak | Top field journal (JOLE, JPubE) |
| Pure theory with no empirical or policy payoff | Weak | TE, JET, RAND |
| Descriptive paper with genuinely new facts/data | Possible | Needs exceptional, first-order novelty |
QJE-vs-siblings taste note: relative to JPE (more sympathetic to a strong calibrated/structural model carrying the paper) and Econometrica (method and theory), QJE leans toward reduced-form, idea-driven empirical work with a memorable lesson. A pure-method paper that AER or JPE might consider for a methods slot is a weak QJE fit.
Sharpening the contribution¶
- Write the one-sentence hook: "We show that [surprising answer] using [clean source of variation]." If you cannot fill both brackets, the project is not ready.
- Identify the broad lesson: finish the sentence "Beyond this setting, our result implies ..."
- Name the debate you move: which prior belief does your answer overturn, quantify, or reconcile?
- Stress-test against desk rejection: QJE desk-rejects fast (~2 weeks). If the editor cannot see the big idea in the first paragraph of the intro, the paper is gone.
Checklist¶
- One-sentence hook with both a surprising answer and a clean source of variation
- A "broad lesson" sentence that travels beyond the specific setting
- The question is legible to a non-specialist top-5 reader
- The answer (not the technique) is what makes the paper interesting
- You can name the prior belief or debate the paper moves
- Not better suited to Econometrica (method) or a top field journal (incremental)
Anti-patterns¶
- A precise causal estimate of something no one doubted ("more education raises wages, again")
- Leading with the method ("we apply the Callaway–Sant'Anna estimator to ...") instead of the question
- A purely descriptive paper with no exceptional, first-order data novelty
- "Interesting only to people who study X" — fails the general-interest filter
- Over-claiming generality the design cannot support (a single-county RDD billed as a universal law)