Qje workflow

QJE Workflow Router (qje-workflow)

Overview

This is the router. It does not replace any specialized skill. It tells you which qje- skill to use at the current stage* of a manuscript aimed at the Quarterly Journal of Economics.

Default assumption: unless the user says otherwise, treat the target as QJE — the oldest English-language economics journal (founded 1886, published by Oxford University Press for the Harvard Department of Economics), a top-5 general-interest journal that rewards a big conceptual idea backed by clean, credible empirics. Its five Harvard-based Editors (Barro, Katz, Nunn, Shleifer, Stantcheva as of 2024) desk-decide in roughly two weeks — the fastest at desk among the top-5 — and desk-reject the majority; unconditional acceptance is only ~1-4%. There is no submission fee; initial submission is a single PDF via Editorial Express under double-blind review. Re-verify volatile specifics (current editors, fee, deposit policy) on the official journal page before relying on them.

When to trigger

• The user asks "what should I do next?"
• The user hands over a draft and needs the current bottleneck diagnosed
• Work is ping-ponging between empirics, theory, writing, and response letters
• A QJE decision letter arrived and the user needs to switch into revision mode

Routing table

Current symptom	Next skill
Idea feels small / "so what?" unclear / not obviously top-5 material	qje-topic-selection
Contribution relative to the literature is fuzzy or undersold	qje-literature-positioning
Empirics rest on OLS + controls; causal claim is undefended	qje-identification
Reduced-form result has no conceptual frame / no model	qje-theory-model
Main result exists but robustness / appendix is thin	qje-robustness
Tables are dense; paper is not figure-forward enough	qje-tables-figures
Prose buries the idea; abstract/intro do not land the "big question"	qje-writing-style
Accepted-stage data/code deposit, or pre-empting referee replication	qje-replication-package
Want to anticipate referee objections before submitting	qje-referee-strategy
Ready to submit via Editorial Express; need a preflight checklist	qje-submission
Received an R&R; need a response-letter strategy	qje-rebuttal

Default order

1. qje-topic-selection — lock the big question + the conceptual takeaway
2. qje-literature-positioning — stake the contribution against the frontier
3. qje-identification — make the causal design bulletproof
4. qje-theory-model — give the result a conceptual frame / model
5. qje-robustness — build the extensive appendix QJE expects
6. qje-tables-figures — finalize figure-forward main exhibits
7. qje-writing-style — make the prose land the idea (abstract + intro last)
8. qje-replication-package — assemble the deposit (also revisit at acceptance)
9. qje-referee-strategy — war-game referee objections pre-submission
10. qje-submission — Editorial Express preflight
11. qje-rebuttal — after the R&R

qje-writing-style is a late-stage polish. Do not rewrite the intro before the identification is settled — the argument will change.

Pipeline status check

Fill this in before asking "what next?" — work top-down and take the first NO as the route.

QJE PIPELINE STATUS — first NO wins
Big question answerable in one sentence a non-specialist cares about?  NO -> qje-topic-selection
Delta vs. the frontier papers named, not just cited?                   NO -> qje-literature-positioning
Causal design survives the toughest referee attack you can imagine?    NO -> qje-identification
Headline coefficient has a conceptual frame or model behind it?        NO -> qje-theory-model
Online appendix already covers the checks referees will demand?        NO -> qje-robustness
Main exhibits figure-forward; no table past ~6 columns?                NO -> qje-tables-figures
Abstract states question, design, and magnitude in plain prose?        NO -> qje-writing-style
QJE Dataverse deposit (data + code) assembled and re-runnable?         NO -> qje-replication-package
Referee objections war-gamed and pre-empted in the draft?              NO -> qje-referee-strategy
Single anonymized PDF ready for Editorial Express (no fee due)?        NO -> qje-submission
Decision letter with an R&R in hand?                                   YES -> qje-rebuttal

Decision shortcuts

• "I have data but no big idea" → qje-topic-selection
• "I don't know who I'm building on or beating" → qje-literature-positioning
• "My DID is TWFE on staggered timing" → qje-identification
• "My result is a coefficient with no story" → qje-theory-model
• "A referee will ask for X robustness" → qje-robustness
• "My main table has 9 columns" → qje-tables-figures
• "The abstract doesn't state the finding" → qje-writing-style
• "Editor wants data and code" → qje-replication-package
• "Submitting tomorrow" → qje-submission
• "Got three referee reports" → qje-rebuttal

Differences vs. other top-5 stacks

If the paper is methods-first (a new estimator, an asymptotic theorem) it belongs at Econometrica; if it is a structural-IO or macro-quantitative paper leading with a calibrated model, JPE or Econometrica may fit better. QJE's comparative advantage is the big empirical-micro question with a clean natural experiment and a broad lesson (labor, public, development, behavioral, economic history, political economy) — the lineage runs from Akerlof's "Market for 'Lemons'" (QJE 1970) and Mankiw–Romer–Weil (QJE 1992) to Chetty–Hendren–Kline–Saez on mobility (QJE 2014). When in doubt, ask whether a smart non-specialist would care about the answer, not the technique. Operational tells that you are at QJE and not a sibling: no submission fee, one PDF via Editorial Express, double-blind review, ~2-week desk decisions, deposit to the QJE Dataverse (not openICPSR).

Anti-patterns

• Do not skip qje-literature-positioning and jump to identification — QJE referees judge the contribution first
• Do not let qje-tables-figures polish exhibits while the identification is still shaky
• Do not let qje-rebuttal draft a response letter before the revised manuscript exists
• Do not treat qje-robustness as optional — a thin online appendix reads as an incomplete paper here