qje-referee-strategy

Referee Strategy (qje-referee-strategy)

When to trigger

• The paper is nearly submission-ready and you want to stress-test it adversarially
• You suspect a specific weakness a referee will seize on
• You are choosing what to put in the body vs. the appendix to forestall objections
• You want to draft suggested/excluded referees and a strategic cover note

The QJE refereeing reality

QJE is the fastest top-5 journal at desk — its five Harvard-based Editors desk-decide in roughly two weeks and desk-reject a clear majority (commonly cited around 60%+), with an unconditional acceptance rate of only ~1-4%. Review is double-blind, so you cannot tailor to a known referee; you must satisfy the toughest plausible expert. Two consequences: (1) the paper has to survive a fast, skim-level editor screen before any referee sees it, so the big idea and clean identification must be visible up front; (2) for papers that survive, QJE grants few rounds with demanding referees, so you usually get one substantive shot. Anticipate the report you would write if you were the toughest expert in your subfield, and answer it in advance.

Map the three referee archetypes

Referee type	What they attack	Pre-empt by...
The identification hawk	The exogeneity claim; pre-trends; exclusion restriction	Falsification + placebo evidence in the body (see qje-identification)
The "so what?" skeptic	Generality; whether the answer matters to all of econ	A crisp broad-lesson + magnitude benchmarking (qje-topic-selection)
The mechanism doubter	Whether the proposed channel is the real one	A testable prediction confirmed in data (qje-theory-model)

War-gaming protocol

1. Pass the desk screen first. Before refereeing even starts, ask: in two pages, is the big question and the clean source of variation obvious? If not, the editor desk-rejects in ~2 weeks regardless of how good the appendix is.
2. Write the hostile report. In 5–8 bullets, list the strongest objections a top expert would raise. Be ruthless; this is cheaper now than after a rejection.
3. Triage. For each: is it a fundamental threat (kills the claim) or a robustness request (adds a table)? Fundamentals must be resolved before submission, not deferred to R&R.
4. Locate the answer. For each objection, point to where in the paper it is already answered. If nowhere, add it (body for fundamentals, appendix for robustness).
5. Decide the framing. Pre-empt the biggest objection explicitly in the text ("A natural concern is X; we address it in three ways ..."). Showing you saw it coming builds referee trust.

Cover-letter / referee suggestions

• Suggest referees who are expert and fair; avoid close collaborators and advisors (conflicts), consistent with the AEA disclosure norms QJE follows.
• Note genuine conflicts to exclude, briefly and professionally.
• Keep any cover note short: the question, the design, the headline result, and fit for a general-interest journal — the editor reading it has ~2 weeks and many submissions.

Checklist

• The two-page editor-skim test passes (big idea + clean variation visible early)
• A written hostile report exists with the 5–8 strongest objections
• Every fundamental threat is resolved in the body before submission
• Each robustness objection has a pre-emptive appendix answer
• The single biggest concern is addressed explicitly and early in the text
• Magnitudes benchmarked so the "so what?" referee is satisfied
• Suggested referees are expert, fair, and conflict-free

Anti-patterns

• Submitting with a known fundamental hole, hoping referees miss it (the editor desk-screen catches a lot first)
• Deferring identification fixes to "future R&R" — at ~1-4% acceptance, QJE may not grant the round
• Ignoring the obvious objection rather than naming and answering it
• Suggesting collaborators/friends as referees (signals naivety or conflict)
• A long, defensive cover letter that argues instead of stating the contribution

Output format

【Desk-screen test】big idea + variation visible in 2 pages? [Y/N]
【Hostile report】[5–8 strongest objections]
【Fundamentals】[threats that must be fixed pre-submission] — status each
【Robustness asks】[appendix answers] — status each
【Pre-empted in text】the biggest concern, addressed where: ...
【Referee suggestions】[expert, fair, conflict-free]
【Next step】qje-submission