JEG's review almost always pairs an economist and a geographer (the board is drawn from both, and editors deliberately route across the bridge). Their objections are systematically different, and a paper that satisfies one often provokes the other. Plan for both, in one manuscript, under double-anonymous review.
| The economist referee tends to attack | The geographer referee tends to attack |
|---|---|
| identification, selection, weak instruments | "where is the geography?" — space as mere fixed effect |
| spatial autocorrelation / overstated inference | absence of mechanism, context, or institutional texture |
| external validity of a single setting | over-formalization that flattens real places |
| whether the model is identified vs. calibrated | a quantitative result with no theory it speaks to |
| magnitude and economic significance | scale/units chosen without geographic reasoning (MAUP) |
For the economist referee: make the identification and inference bulletproof before submission — modern staggered estimators, Conley SEs, spillover bounds, a defended instrument (see jegeo-identification, jegeo-robustness). Report magnitudes, not asterisks. Address external validity head-on.
For the geographer referee: make place do real analytical work — name the mechanism in prose, justify the spatial scale substantively, engage the evolutionary/institutional literature, and ensure the maps argue (see jegeo-theory-model, jegeo-tables-figures). Never let space be only a control.
The bridge insurance: the single best defense is a paper where the mechanism is the same object both referees can grip — the economist sees it identified, the geographer sees it grounded in place. If you can write that one sentence, you can usually satisfy both.
Under double-anonymous review the cover letter is your one chance to tell the editor how to balance the panel. Use it deliberately:
A cover letter that frames the paper as single-community invites a single-community desk read, which at JEG is a fast rejection.
A paper using a quantitative-spatial model to evaluate a transport investment is strong on identification and inference. The author, an economist, armors only that side. Anticipate the geographer referee: they will ask why "region" is the unit (MAUP), where the institutional mechanism is (why do firms actually relocate?), and whether the model flattens real places into representative agents. Pre-empting that side before submission — adding a commuting-zone robustness check, a prose mechanism, and a paragraph on heterogeneity across regional contexts — converts a likely split decision into a clean R&R. The economist's questions were already answered; the unanticipated geographer was the real risk.
Read your own paper twice, once as each referee:
Any "no" in either pass is a risk to fix now or pre-answer in the text — not to discover in the reports.
【Journal】Journal of Economic Geography
【Skill】jegeo-referee-strategy
【Economist-side risks】top objections + fix status (now/anticipated/R&R)
【Geographer-side risks】top objections + fix status (now/anticipated/R&R)
【Bridge sentence】the one mechanism both referees can grip
【Reviewer suggestions】names/areas spanning both communities
【Anonymity】double-anonymous clean? [Y/N]
【Next skill】jegeo-submission