Literature Positioning (jet-literature-positioning)

When to trigger

• Writing the related-work paragraph(s) for a JET theory paper
• A referee or editor may ask "how is this different from [closest existing theorem]?"
• You need to defend originality at the desk screen

How JET wants the literature handled

JET spans general theory, so referees are expert in the nearest existing results, not a broad empirical literature. Positioning is theorem-relative, not topic-relative. For each closest paper, state the delta in one of these precise forms:

• Weaker assumptions: "X obtains the result under [strong condition]; we obtain it dropping/weakening it to [weaker condition]."
• Greater generality: "their characterization holds for [finite / quasi-linear / single-crossing] environments; ours covers [the general case]."
• New object: "no prior result characterizes [the equilibrium set / optimal mechanism / preference domain] in this environment; Theorem 1 does."
• Tighter / constructive: "existence was known; we give the [closed form / sharp bound / algorithm]."

Reference-handling rules (JET-specific)

• References cited in the abstract must be written out in full — so cite sparingly there.
• Keep unpublished results and personal communications out of the reference list; lean on published theorems as the comparison set.
• First submission can use any complete, consistent reference format; JET's proof-stage style is name-year / author-year, so elsarticle-harv is the safest LaTeX default.

Theorem-delta table

Build a private table before writing related work:

Only the strongest two or three rows belong in the manuscript. The table prevents vague "we extend" language and makes overclaiming visible before a referee catches it.

Subfield frontier maps

The comparison set differs by JET area; find the closest theorem inside the right lineage, name the lineage in one clause, then jump straight to the single nearest result:

• Mechanism design / auctions: the Myerson optimal-auction and implementation lineage; typical deltas weaken the common prior, risk neutrality, or transferable utility.
• Matching / market design: the Gale–Shapley stability and Kelso–Crawford substitutes line; deltas relax substitutability or add contracts, constraints, or distributional objectives.
• Decision theory / ambiguity: Anscombe–Aumann-based axiomatizations (maxmin, variational, smooth ambiguity); deltas trade one axiom for a behaviorally weaker one.
• Information design / persuasion: the Bayesian-persuasion concavification line; deltas add senders, dynamics, or robustness to the receiver's beliefs.
• Repeated games / dynamic contracts: folk-theorem and recursive-contract traditions; deltas alter monitoring, commitment, or the discounting structure.

Positioning-paragraph skeleton

The closest result is [Author (Year), Theorem k], which proves [object] under [assumption set S].
Our Theorem 1 [weakens S to S' / covers the general (non-quasi-linear / infinite-type) case /
characterizes an object their analysis leaves open]. The techniques also differ: their argument
relies on [tool]; ours requires [new tool] because [what breaks under the weaker assumptions].
Relative to the applied literature on [topic], our contribution is the theorem itself, not a
new application of existing results.

Vignette: placing an information-design result

A hypothetical paper characterizes optimal disclosure when the receiver is maxmin. Two candidate anchors compete: the standard persuasion concavification theorem (delta: receiver ambiguity breaks Bayesian updating, so the sender's value is no longer a concavification) and the robust-mechanism literature (delta: the designer commits to information rather than transfers). Write both rows in the theorem-delta table, then lead the related-work paragraph with the anchor whose assumptions you actually weaken — the other becomes one supporting sentence, not a co-headline.

Anti-patterns

• A multi-page survey that never states your delta against the single closest theorem
• "We contribute to the literature on X" with no assumption- or generality-level comparison
• Citing working papers as the frontier when a published theorem already settles the comparison
• Vague novelty ("we extend the model") instead of naming the relaxed assumption

Output format

【Closest result】<author, year, the theorem>
【Our delta】weaker assumption | greater generality | new object | tighter/constructive
【Stated as】"<one-sentence positioning to put in the intro>"
【Abstract cites】kept minimal + written in full? [Y/N]
【Next】jet-identification-strategy (assumptions & proof) / jet-contribution-framing