Jde review process

Review Process (jde-review-process)

When to trigger

• You are deciding among the full-length, short-paper, and pre-results review routes
• You need realistic expectations about JDE review and timelines
• You have a prospective design and want it reviewed before results exist
• You want to confirm you are within the submission cap

How JDE review works

JDE is the leading field journal in development economics (Elsevier), edited by Editors-in-Chief A. Foster and K. Macours, with Dean Karlan among the Co-Editors. Review is single-anonymized — referees know the author's identity; the author does not know the referees. It is highly selective: ~1,300 submissions a year, only about a quarter sent for review, and roughly 6-8% published. Submission is via Editorial Manager.

There is a USD 50 non-refundable submission fee for original manuscripts. Open access is optional; the current APC is USD 4,760 excluding taxes.

Submission cap: under normal circumstances an author may not submit more than three papers within any 12-month period. Plan submissions accordingly.

The three routes

Full-length (standard)

The default track. No fixed length limit; an extensive online appendix is expected. Full single-anonymized review with multiple referees and R&R rounds.

Short-paper (limited-revision) track

Modeled on AER: Insights: at most 6,000 words, at most 5 exhibits (tables or figures), and an online appendix of at most 20 pages (the manuscript must be self-contained). Refereed decisions generally occur within 6-8 weeks and use a limited-revision structure. Best for a sharp, self-contained result that does not need a long treatment.

Pre-results review (Registered Reports)

A permanent track (piloted 2018-2019 with BITSS) for prospective empirical projects. You submit a Stage 1 proposal — background/motivation, hypotheses, study procedures, a statistical analysis plan, a power analysis, and pilot data if applicable, with a cover page carrying JEL codes and ethics/funding/conflict disclosures. The Stage 1 proposal must not exceed 60 pages and its abstract is capped at 150 words. If accepted in principle before results are known, you implement the study and submit the Stage 2 manuscript. Published pre-results papers look like any other JDE article except for a footnote noting pre-results submission and the inclusion of the pre-specified plan in the supplementary online appendix. This route reduces publication bias and is well suited to RCTs and other prospective designs.

Choosing a route

• Sharp single result, self-contained → short-paper track
• Prospective design, want the plan locked and de-risked → pre-results review
• Rich result needing full treatment, theory, or extensive empirics → full-length

Route-choice decision table

Your situation	Route	Why
RCT not yet run; want the design locked and bias-proofed	Pre-results	Stage 1 accepted in principle before results exist
One sharp result, ≤ 6,000 words, ≤ 5 exhibits, self-contained	Short-paper	Fast (~6–8 wk), limited revision, editor decides
Rich design + theory + extensive heterogeneity and robustness	Full-length	No length cap; extensive online appendix expected
Quasi-experiment already estimated, needs full empirical defense	Full-length	Multiple referees, multiple R&R rounds

Worked routing example (illustrative)

Hypothetical: a team has designed but not yet fielded a cluster-randomized cash-transfer trial in a low-income setting and wants to de-risk it.

• Right call: Stage 1 pre-results submission — background, hypotheses, procedures, a statistical analysis plan, and a power analysis at the cluster level (e.g., 90 clusters powered to detect a 0.15 SD consumption effect, illustrative). If accepted in principle, results cannot sink the paper for being null.
• Wrong call: running the trial first, finding a null, then trying pre-results — that track is prospective only.

Calibration anchors (hedged)

• Selectivity is high: ~1,300 submissions/year, about a quarter sent for review, ~6-8% published.
• Short-paper refereed decisions generally land in 6-8 weeks; the full-length pipeline runs months across rounds — plan the three-per-12-months cap around it.
• The development-policy-relevance bar is real: a technically clean paper with no first-order LMIC stake is a routine non-starter regardless of route.

Anti-patterns

• Submitting a sprawling paper to the short-paper track (it caps words and exhibits)
• Treating pre-results review as a place to submit after you already have results
• Exceeding three submissions in a 12-month window
• Assuming double-blind anonymization — JDE review is single-anonymized

Output format

【Route】full-length / short-paper / pre-results review
【Why this route】...
【Within 3-per-12-months cap?】[Y/N]
【Route-specific limits met】words/exhibits/pages as applicable
【Next step】jde-submission