R&R Rebuttal (jeg-rebuttal)

When to trigger

• A JEG revision decision has arrived
• Referees request stronger growth framing, theory, empirics, or calibration
• You need a response letter and aligned manuscript changes

Triage

1. Fit and contribution: growth mechanism not clear enough.
2. Theory: assumptions, equilibrium, dynamics, comparative statics, or proofs.
3. Empirics: identification, sample, robustness, historical measurement, or external validity.
4. Calibration/simulation: parameter sources, targeted moments, sensitivity, transition paths.
5. Policy and files: data availability, declarations, source files.

Response strategy

• Lead with the editor's synthesis and the highest-value changes.
• Add quantitative sensitivity where a calibration concern is central.
• Add comparative estimates where empirical growth literature raises conflicting results.
• Narrow claims when model or data scope is local.
• Update source files and declarations alongside the manuscript.

Revision evidence map

For each major referee concern, identify the evidence type:

• Mechanism concern -> new model intuition paragraph, proposition, or transition-path figure.
• Assumption concern -> relaxed-assumption result, proof clarification, or sensitivity table.
• Empirical concern -> identification diagnostic, alternative sample, or comparative estimate.
• Calibration concern -> parameter-source table plus one-at-a-time and joint sensitivity.
• Data concern -> Data Availability Statement, code archive, source-file update, or access note.

The response letter should show which evidence changed the manuscript, not just where text was added.

JEG referee objection playbook

Map the objections most common at this journal to their accepted repairs:

• "Spatial autocorrelation inflates your significance" → recompute Conley SEs at several distance cutoffs, add a spatially correlated placebo distribution, and put both in the revised table notes — not only in the letter.
• "The exclusion restriction is implausible" → confront each published alternative channel for the instrument, control for those channels directly, and add an over-identification or bounding exercise rather than restating the assumption with more conviction.
• "Persistence with no mechanism" → bring intermediate-period outcomes, a decomposition, or a model section, and promote the mechanism evidence to a main exhibit.
• "Calibration moments look cherry-picked" → add untargeted-moment validation plus one-at-a-time and joint sensitivity; tabulate every parameter's source.
• "The country sample is too selective for a growth claim" → expand or re-weight the sample and bound the claim's scope explicitly in abstract and conclusion.
• "This is not about growth" → reframe around divergence, convergence, or transition dynamics; growth editors rarely accept a fit defense made by assertion.

Worked vignette — answering a spatial-inference objection

Illustrative R&R: Referee 2 argues the headline t-statistic of 4.8 on a persistence coefficient is an artifact of spatial correlation. The response that works:

1. Recompute Conley SEs at 100/250/500 km: t falls to 3.6/3.2/3.1 — significance survives; all three appear in the revised Table 2 notes.
2. Generate 1,000 spatial-noise placebo treatments with a matched correlation range; the true coefficient sits above the 99th percentile of the placebo distribution (illustrative).
3. Add macro-region fixed effects; the coefficient moves from 0.21 to 0.18, inside the original confidence band.
4. Letter structure: quote the concern, give the three numbers, point to the exact revised table and figure, and state what changed in the text — two paragraphs, zero defensiveness.

Letter discipline for this venue

• When concerns overlap across referees, order responses by growth substance — fit, mechanism, identification, calibration — and cross-reference instead of duplicating analyses under each referee number.
• Quantify every concession: if the claim is narrowed, show the revised abstract sentence verbatim.
• Fold every new analysis into the replication package and say so in the letter; at a venue this data-disciplined, revision credibility extends to the artifacts.

Output format

[Decision] major / minor
[Main growth concern] ...
[New analyses/proofs] ...
[Sensitivity/robustness] ...
[Manuscript locations] ...
[Next step] resubmit through Springer portal