Topic Selection (jue-topic-selection)

When to trigger

• The question is interesting but it is unclear it is spatial — could it be a labor or public-finance paper that merely uses geographic data?
• The paper sits on a boundary with RSUE, Journal of Regional Science, JEG (OUP), or JPubE and you must decide where it belongs
• A reviewer or coauthor asks "why is this urban economics?" and the answer is thin
• You have a clean result but no urban mechanism — only a geographic correlation
• You are choosing between a full JUE article and the short-paper JUE: Insights track

Is this a JUE paper? The spatial-mechanism test

JUE publishes work where space is constitutive of the economics, not a label on the data. The question should turn on at least one spatial primitive: agglomeration economies, housing supply elasticity, land use & zoning, commuting/transportation, local public finance & Tiebout sorting, neighborhood effects, or spatial sorting/equilibrium. Run the manuscript through three gates:

1. Mechanism gate. Can you name a spatial mechanism — capitalization, agglomeration spillover, sorting across locations, congestion, market access — that generates the result? If the mechanism is a-spatial (it would hold with units relabeled non-geographically), it is not a JUE paper.
2. Equilibrium gate. Does the question respect that people and firms move? A partial result that ignores spatial reallocation (workers/firms/residents re-sorting) will draw the JUE referee's first objection. The contribution should engage spatial equilibrium even if reduced-form.
3. Marginal-contribution gate. Urban economics is mature; "X affects house prices / city size / wages" is rarely enough. The novelty must be a new mechanism, a new spatial design, novel geocoded data, or a quantitatively important magnitude for urban policy.

Sibling-boundary decision table

Choosing the article format

• Full JUE article — a complete contribution: design, mechanism, magnitudes, robustness, policy reading.
• JUE: Insights — a focused, well-identified short result. Cap ≤6,000 words and ≤5 exhibits (each exhibit below five buys +200 words; a zero-exhibit paper ≤7,000 words) (检索于 2026-06；以官网为准). Use it for a single sharp finding (a clean policy evaluation, a measurement contribution) that does not need a full model. The track was launched to publish timely, tightly-identified short papers; do not pad a Insights-sized result into a full article to look weightier.

What gets desk-rejected at JUE

JUE is a high-volume field flagship and the editors desk-reject freely. The recurring desk-reject patterns are worth designing against from the start:

• No spatial mechanism — a result that uses geographic units but whose economics is a-spatial (a labor or finance result that happens to be measured by city).
• Setting mistaken for contribution — "the first study of [city/country]" with no new mechanism, design, data, or magnitude.
• Wrong venue — a tax-design paper (JPubE), a spatial-method paper (RSUE), or an economic-geography paper (JEG) sent to JUE because it is "about cities."
• Identification not credible for space — a cross-location OLS that any urban referee will read as sorting, presented as causal.
• Stale question with no new angle — re-confirming a well-established urban regularity without moving the frontier.

Timeliness and the JUE: Insights niche

The Insights track was built for results that are sharp and timely — a clean evaluation of a recent policy, a measurement contribution, a decisive null. If your finding is one well-identified estimate that does not need a model and would lose force if delayed by a multi-year full-article cycle, Insights is the right home. Reserve the full article for contributions that need a model, a battery of mechanisms, or a quantitative-spatial counterfactual. Misjudging this — padding an Insights result or compressing a full contribution — is itself a fit error referees notice.

Worked vignette (illustrative)

A team has clean admin data showing firms near a new university campus pay higher wages. Tempting framing: "agglomeration raises wages." But the mechanism gate exposes a sorting story — high-wage firms may have located near the campus. The JUE-viable reframing pins a mechanism: use the staggered campus openings as a shock to local knowledge spillovers, decompose the wage gain into sorting vs true spillover, and report the spillover magnitude relevant to place-based innovation policy. That is an agglomeration contribution; "firms near campuses pay more" is a desk-reject.

Checklist

• A spatial mechanism is named in one sentence, and the result would not survive relabeling units non-geographically
• The paper engages spatial equilibrium / reallocation rather than assuming a closed location
• The marginal contribution is more than "X affects an urban outcome"
• The sibling boundary (RSUE / JRS / JEG / JPubE / AEJ:Applied) is explicitly resolved
• Format chosen (full article vs JUE: Insights) matches the scope of the result
• The urban-policy or theory audience who would cite this paper is identifiable
• The spatial resolution the design needs exists and is accessible (and depositable)
• The result is not a known urban regularity re-confirmed with no new angle

Anti-patterns

• A national/firm-level result run on county or city data with no spatial mechanism ("geographic data ≠ urban economics")
• Ignoring that agents re-sort, so the estimated effect double-counts or omits general-equilibrium reallocation
• Pitching to JUE a paper whose real contribution is a tax design (→ JPubE) or a spatial-econometrics method (→ RSUE)
• Forcing a thin result into a full article when JUE: Insights fits better, or vice versa
• Claiming "first paper to study X city" as if novelty of setting were novelty of contribution

Data-availability shapes feasibility — decide early

Urban questions are often constrained by what geography you can observe. Before committing, confirm the spatial resolution you need actually exists and is accessible: parcel/address-level for capitalization and boundary designs, establishment-level for agglomeration, tract/block for neighborhood effects, network/GTFS for transport. A question that requires geocoded individual data you cannot obtain (or cannot deposit under JUE's replication policy) is a topic-selection problem, not a later logistics detail — fold the data path and the replication route into the topic decision, and loop in jue-replication-package if the data is restricted.

Output format

【Spatial mechanism】one sentence — what space does in the economics
【JUE vs sibling】JUE / RSUE / JRS / JEG / JPubE / AEJ:Applied + one-line reason
【Format】full article / JUE: Insights
【Marginal contribution】mechanism / design / data / magnitude (pick the real one)
【Equilibrium engagement】how reallocation is handled
【Next skill】jue-literature-positioning

If the spatial-mechanism test fails, the fastest fix is rarely a new dataset — it is reframing the existing result around the spatial primitive (capitalization, sorting, agglomeration) that was always implicit but never made the point of the paper.