JEL Classification System (jel-classification-system)

When to trigger

• The survey is drafted and you must assign JEL classification codes for submission
• You are unsure how many codes, at what level, or in what order to list them
• You want the survey's scope to map cleanly onto the JEL taxonomy a reader uses to find it
• You are reasoning about which JEL category the field you are surveying actually sits in

What the JEL classification system is (and why JEL owns it)

The JEL classification system is the American Economic Association's subject taxonomy for economics, maintained by the Journal of Economic Literature itself and used across EconLit and most economics journals. A survey for JEL therefore has a dual relationship with the codes: it must be indexed by them like any paper, and — because JEL is the system's steward — a well-scoped survey often corresponds to one or a small set of categories that define its field.

Structure (检索于 2026-06；以官网为准):

• Three levels. A single letter (20 top categories), a two-digit subcategory (e.g. J2), and a three-character detailed code (e.g. J24).
• Assign detailed three-character codes for a submission; the letter/two-digit levels locate the field.

The 20 top-level categories:

Assigning codes to a survey

1. Find the field's home category first (the letter), then the subcategory, then detailed codes — top-down, so you do not miss the natural primary code.
2. Lead with the field's primary detailed code, then add codes for the methods and adjacent fields the survey substantively covers (a survey is broader than a paper, so a handful of codes is normal).
3. Mirror the organizing framework. The codes should look like the survey's cells: if a section covers an econometric method the field relies on, include the relevant C code; if it spans into public economics, include the H code.
4. Cross-check against the EconLit code list for the exact current label and three-character code — labels and codes are revised; do not assign from memory (检索于 2026-06；以官网为准).
5. Confirm count/format limits for the manuscript in the AEA JEL style guide via jel-submission (JEL codes are commonly listed with keywords on the title page).

Because JEL maintains the classification system, a survey is in an unusual position: it can illuminate where a field sits in the taxonomy, and occasionally a maturing field's survey is part of the case that a code's scope should be reconsidered. You are not assigning codes from outside the system — you are working inside the AEA's own subject map, so get the codes exactly right.

Checklist

• Primary detailed (three-character) code identified top-down (letter → two-digit → detailed)
• Method codes (e.g. C) and adjacent-field codes added where the survey substantively covers them
• Code set mirrors the organizing framework's cells (not a random scatter)
• Each code verified against the current EconLit JEL code list (label + three-character form)
• Primary code listed first; codes ordered by centrality to the survey
• Count/format limits confirmed in the AEA JEL style guide (via jel-submission)
• Not over-coding: only categories the survey genuinely covers

Anti-patterns

• Assigning only a single broad letter-level code to a wide-ranging survey (under-indexing)
• Scattering a dozen codes for fields the survey barely touches (over-indexing)
• Inventing or mis-remembering three-character codes instead of checking the live EconLit list
• Codes that contradict the organizing framework (the survey's own map of its scope)
• Treating the codes as an afterthought rather than a true description of the field's place in economics

Output format

【Field home】<letter> → <two-digit> → <primary detailed code>
【Code set】<primary code first, then method + adjacent-field codes>
【Framework mirror】codes match the survey's cells? Y/N
【Verification】each code checked on current EconLit list? Y/N
【Format】count/placement confirmed in AEA JEL style guide? Y/N
【Next step】→ jel-editor-strategy / jel-submission (codes go on the title page with keywords)