Accessibility & Translation (jep-accessibility-and-translation)
When to trigger
- The draft reads like a working paper with the proofs deleted
- A labor or macro economist who doesn't work on the topic gets lost
- Sentences carry notation, estimator names, or subfield jargon a generalist won't know
- You worry that simplifying will make the piece sound shallow
The JEP accessibility bar
JEP articles should be readable by 90 percent or more of the AEA membership and useful to economists not conversant with the author's subspecialty; the journal is "scholarly without relying too heavily on mathematical notation," explaining ideas through economic intuition rather than derivations (检索于 2026-06；以官网为准). Translation is not simplification of the ideas — it is removing the barriers (notation, jargon, insider shorthand) that hide the ideas from a smart non-specialist.
Translation moves (technical → JEP)
| Technical draft |
JEP move |
| An estimating equation in the body |
State what the equation says in words; the design goes to a note or appendix if at all |
| "We instrument X with Z" |
"We exploit [the natural experiment] so that the variation in X is plausibly unrelated to [confounder]" |
| A Greek-letter parameter |
Name the economic quantity it stands for and give its sign/magnitude in plain units |
| Subfield jargon ("the intensive margin," "MPC," "ATT") |
Define on first use in a clause, or replace with plain words where precision allows |
| "Statistically significant at 1%" |
A number with a plain sense of precision (see jep-evidence-without-equations) |
| A model's machinery |
The economic mechanism — the story of what drives what — first; the formalism is optional scaffolding |
Principles
- Intuition first, formalism optional. Lead with the mechanism in words. If notation appears at all, it follows the intuition and could be skipped without losing the thread.
- Define on first use, sparingly. Every necessary term gets a one-clause gloss the first time. Better: avoid the term.
- Concrete over abstract. A vivid example or a number anchors an abstract claim.
- Don't condescend. The reader is a professional economist, just not in your field — explain the subfield-specific, assume the general.
- Cut insider signaling. Citations-as-name-dropping, method-flexing, and "as is well known" add nothing for a generalist.
Calibrating depth (not dumbing down)
The test is not "could a layperson read this" but "could a non-specialist economist follow the argument and trust it." Keep the rigor of the reasoning; remove the rigor-display. If a precise technical point is essential, state it plainly and put the formal version in a footnote or short appendix — the body must stand on its own.
Checklist
Worked vignette (illustrative)
A working-paper sentence reads: "We identify β via a shift-share instrument, exploiting Bartik-style variation in local labor demand, and recover an IV estimate that survives the Borusyak–Hull exogeneity diagnostics." The JEP translation: "To separate cause from coincidence, we lean on the fact that some towns were more exposed to a national hiring boom simply because of the industries they happened to host — exposure that had little to do with anything local. Comparing more- and less-exposed towns lets us read off how the boom changed wages." The mechanism and the credibility are now legible to any economist; the estimator name and diagnostics, if needed at all, go to a footnote.
Anti-patterns
- "Translating" by deleting the math but keeping the jargon — still unreadable
- Dumbing down the ideas instead of removing the barriers (loses rigor and respect)
- Leaving one estimating equation "for the experts" in an otherwise accessible body
- Defining nothing because "everyone knows this" — they don't, across fields
- Over-simplifying into vagueness so the claim no longer means anything checkable
Output format
【Jargon/notation removed】[list of terms translated or cut]
【Mechanism in plain words】[one paragraph]
【Equations/proofs relocated】[to note / appendix / cut]
【Non-specialist read test】can a general economist restate it? [Y/N]
【Rigor preserved?】reasoning intact, only display removed? [Y/N]
【Next step】jep-evidence-without-equations
1---2name: jep-accessibility-and-translation3description: Use when translating technical results, models, and methods into plain language for a Journal of Economic Perspectives (JEP) article — minimizing jargon and notation without dumbing down. Handles language and translation of ideas; defer evidence/number presentation to jep-evidence-without-equations and exhibits to jep-exhibits-for-general-readers.4---56# Accessibility & Translation (jep-accessibility-and-translation)78## When to trigger910- The draft reads like a working paper with the proofs deleted11- A labor or macro economist who doesn't work on the topic gets lost12- Sentences carry notation, estimator names, or subfield jargon a generalist won't know13- You worry that simplifying will make the piece sound shallow1415## The JEP accessibility bar1617JEP articles should be readable by **90 percent or more of the AEA membership** and useful to economists **not conversant with the author's subspecialty**; the journal is "scholarly without relying too heavily on mathematical notation," explaining ideas through **economic intuition** rather than derivations (检索于 2026-06；以官网为准). Translation is **not** simplification of the *ideas* — it is removing the *barriers* (notation, jargon, insider shorthand) that hide the ideas from a smart non-specialist.1819## Translation moves (technical → JEP)2021| Technical draft | JEP move |22|-----------------|----------|23| An estimating equation in the body | State what the equation *says* in words; the design goes to a note or appendix if at all |24| "We instrument X with Z" | "We exploit [the natural experiment] so that the variation in X is plausibly unrelated to [confounder]" |25| A Greek-letter parameter | Name the *economic quantity* it stands for and give its sign/magnitude in plain units |26| Subfield jargon ("the intensive margin," "MPC," "ATT") | Define on first use in a clause, or replace with plain words where precision allows |27| "Statistically significant at 1%" | A number with a plain sense of precision (see `jep-evidence-without-equations`) |28| A model's machinery | The *economic mechanism* — the story of what drives what — first; the formalism is optional scaffolding |2930## Principles3132- **Intuition first, formalism optional.** Lead with the mechanism in words. If notation appears at all, it follows the intuition and could be skipped without losing the thread.33- **Define on first use, sparingly.** Every necessary term gets a one-clause gloss the first time. Better: avoid the term.34- **Concrete over abstract.** A vivid example or a number anchors an abstract claim.35- **Don't condescend.** The reader is a professional economist, just not in your field — explain the *subfield-specific*, assume the *general*.36- **Cut insider signaling.** Citations-as-name-dropping, method-flexing, and "as is well known" add nothing for a generalist.3738## Calibrating depth (not dumbing down)3940The test is not "could a layperson read this" but "could a *non-specialist economist* follow the argument and trust it." Keep the rigor of the *reasoning*; remove the rigor-*display*. If a precise technical point is essential, state it plainly and put the formal version in a footnote or short appendix — the body must stand on its own.4142## Checklist4344- [ ] No estimating equations or proofs in the body (intuition in words instead)45- [ ] Every subfield term defined on first use or replaced46- [ ] Each Greek/parameter symbol replaced by the economic quantity it denotes47- [ ] A non-specialist economist can restate the main mechanism after one read48- [ ] No method-flexing or citation name-dropping49- [ ] Necessary technical precision moved to a note/appendix, body self-contained50- [ ] Simplified language did not silently weaken or overstate a claim (cross-check `jep-balance-and-objectivity`)5152## Worked vignette (illustrative)5354A working-paper sentence reads: "We identify β via a shift-share instrument, exploiting Bartik-style variation in local labor demand, and recover an IV estimate that survives the Borusyak–Hull exogeneity diagnostics." The JEP translation: "To separate cause from coincidence, we lean on the fact that some towns were more exposed to a **national hiring boom** simply because of the industries they happened to host — exposure that had little to do with anything local. Comparing more- and less-exposed towns lets us read off how the boom changed wages." The mechanism and the credibility are now legible to any economist; the estimator name and diagnostics, if needed at all, go to a footnote.5556## Anti-patterns5758- "Translating" by deleting the math but keeping the jargon — still unreadable59- Dumbing down the *ideas* instead of removing the *barriers* (loses rigor and respect)60- Leaving one estimating equation "for the experts" in an otherwise accessible body61- Defining nothing because "everyone knows this" — they don't, across fields62- Over-simplifying into vagueness so the claim no longer means anything checkable6364## Output format6566```67【Jargon/notation removed】[list of terms translated or cut]68【Mechanism in plain words】[one paragraph]69【Equations/proofs relocated】[to note / appendix / cut]70【Non-specialist read test】can a general economist restate it? [Y/N]71【Rigor preserved?】reasoning intact, only display removed? [Y/N]72【Next step】jep-evidence-without-equations73```
brycewang-stanford