In economics, the working paper is not a minor pre-publication courtesy—it is the field’s primary mode of scholarly communication. Findings routinely circulate as a working paper or discussion paper years before (or instead of) appearing in a peer-reviewed journal, and a paper’s working-paper number is often the citation a researcher’s CV, grant application, or job-market packet leans on. Three series dominate this landscape for economists: the National Bureau of Economic Research (NBER) Working Paper Series, the IZA Institute of Labor Economics Discussion Paper series, and the Centre for Economic Policy Research (CEPR) Discussion Paper series. This guide covers how researchers get papers into each series, the conventions that set economics working papers apart from a standard journal submission, and why the working-paper stage matters so much for dissemination and priority-claiming in this discipline.
Why economics runs on working papers
Formal peer review in economics is slow—multi-year turnaround from first submission to publication is common at top journals. Working-paper series exist to close that gap: they let researchers put findings in front of the field, establish a public record of when a result was produced, gather informal feedback, and accumulate citations, all before (and often regardless of) formal journal acceptance. A paper’s position in a working-paper archive functions as a de facto priority claim—the dated, citable, publicly indexed version of record for “who found this first”—independent of how long the eventual journal review process takes. This is why economics job-market candidates, tenure files, and grant reports routinely cite NBER, IZA, or CEPR numbers directly rather than waiting for a journal citation.
This differs structurally from a preprint server like arXiv or SSRN: NBER, IZA, and CEPR are not open, self-deposit repositories. Each is an institutional series gated by affiliation, and each imposes its own front-matter, disclaimer, and formatting conventions rather than accepting an author’s manuscript as-is.
NBER Working Paper Series
The National Bureau of Economic Research, a private, nonprofit U.S. economic research organization, distributes more than 1,200 working papers a year, describing them as new research by NBER affiliates circulated prior to peer review for discussion and comment.
- Who can submit: Authorship is restricted to NBER-affiliated researchers—NBER faculty research associates, faculty research fellows, and similar formal affiliates. A researcher without an NBER affiliation cannot deposit a paper directly; papers reach the series through an affiliated co-author or through the relevant NBER research program.
- Review before circulation: Papers are circulated prior to formal peer review, not after—the series is explicitly a pre-peer-review dissemination channel, not a certified/refereed output. NBER states its papers “may not offer policy recommendations or normative judgments about policies,” though they may report analytic results on the effects of policies.
- Access tiers: Working papers become open access roughly 18 months after release; before that, access is limited to NBER-affiliated academics, corporate associates, U.S. government employees, journalists, and researchers in low-income countries.
- Format: Each paper is assigned a sequential NBER working paper number and carries a standard NBER cover page (title, authors, program affiliation, date, and the standard NBER disclaimer that views are those of the authors and not the Bureau).
IZA Discussion Papers
The IZA Institute of Labor Economics (formerly the Forschungsinstitut zur Zukunft der Arbeit) runs a discussion paper series—now coordinated through the Luxembourg Institute of Socio-Economic Research (LISER)—that has grown to more than 17,000 papers and is one of the most heavily used dissemination channels in labor economics specifically.
- Who can submit: The series is described as carrying research output “by IZA staff and network members,” meaning submission depends on being IZA staff or a member of the IZA research network (IZA Research Fellow/Affiliate), not open self-submission from any economist.
- Purpose: IZA positions the series explicitly as making new research accessible before it appears in refereed journals—the same pre-journal dissemination role NBER and CEPR play, but scoped to labor economics and closely related fields (family economics, migration, education economics, and similar).
- Discoverability: Papers are indexed and searchable by title, abstract, keywords, JEL classification code, author, and paper number, and IZA offers subscription alerts for newly released papers in a researcher’s area.
CEPR Discussion Papers
The Centre for Economic Policy Research, a European network of economists, publishes over 1,000 discussion papers a year through its own series, with an archive of more than 20,000 papers.
- Who can submit: Research entering the CEPR Discussion Paper series comes from CEPR Research Fellows and Research Affiliates—formal, appointed positions within the CEPR network. As with NBER, this is not an open-submission series; a researcher generally needs an existing CEPR affiliation (or a co-author who holds one) to place a paper in the series. CEPR’s Research Fellow and Affiliate appointments are themselves periodically opened for nomination.
- Purpose of circulation: CEPR describes discussion papers as circulated widely to the research and policy community specifically so the results receive prompt, thorough professional scrutiny—an explicit acknowledgment that the series substitutes informal, rapid community feedback for the slower formal referee process at this stage.
Conventions distinct from journal submission
Preparing a paper for a working-paper series differs from preparing the same paper for journal submission in several concrete ways:
- No blind review, no anonymization. Author names, affiliations, and acknowledgments appear on the cover page from the outset—there is no double- or single-anonymized review to prepare a manuscript for, unlike a typical journal submission.
- Institutional cover page and disclaimer. Each series has its own standard front matter: a series logo, a sequential paper number, and (for NBER in particular) a standard disclaimer that the views expressed are those of the authors and do not represent the position of the institution.
- JEL classification codes. Economics working papers conventionally list Journal of Economic Literature (JEL) classification codes on the cover page or abstract, a discipline-specific convention with no direct equivalent in most other fields’ preprint practice.
- Living documents. Unlike a journal submission, which is a fixed snapshot reviewed by editors, a working paper is frequently revised in place after release—authors commonly post updated drafts under the same paper number as they respond to informal feedback, well before (or instead of) a formal journal revise-and-resubmit cycle.
- No length or venue constraints from a target journal. Because the paper isn’t yet formatted for a specific journal’s house style, working-paper drafts are often longer and more heavily appendiced than the eventual published version, with additional robustness checks, proofs, or data description moved into a working-paper-only appendix.
- Parallel indexing. Papers in all three series are typically also indexed on SSRN and in RePEc/IDEAS, the economics discipline’s dedicated bibliographic database, which is how citation counts accrue to a working paper even before journal publication.
From working paper to journal publication
None of the three series above is a substitute for journal publication, and none claims to be—each explicitly frames itself as pre-peer-review circulation. A typical trajectory is: draft circulates as a working paper, the author collects seminar and informal referee-style feedback, the paper is substantially revised (sometimes repeatedly, under the same working-paper number), and only then submitted to a journal, where it enters formal peer review independent of its working-paper history. Because the working paper and the eventual published article can differ substantially after review, researchers and administrators tracking a body of work should treat the working-paper version and the journal version as related but distinct outputs, both worth citing correctly according to which version informed the citing work.
Frequently asked questions
Do I need to be affiliated with NBER, IZA, or CEPR to get a working paper into their series?
Yes, in all three cases. These are not open-submission preprint servers. NBER working papers require NBER affiliation (faculty research associate, faculty research fellow, or similar formal status); IZA discussion papers require being IZA staff or an IZA research network member; CEPR discussion papers require being a CEPR Research Fellow or Research Affiliate. A researcher without one of these affiliations typically needs a co-author who holds it.
Are NBER, IZA, and CEPR working papers peer reviewed?
No, not at the point of release into the series. All three explicitly describe their series as pre-peer-review dissemination channels. CEPR frames this as intentional: wide circulation is meant to generate rapid, informal community scrutiny before or in parallel with the slower formal journal referee process.
What’s the difference between a working paper and a discussion paper?
In practice, the terms are used interchangeably across these series—NBER calls its series “working papers,” while IZA and CEPR call theirs “discussion papers.” Both refer to the same kind of output: a numbered, citable, institutionally issued draft circulated before or independent of formal journal publication. See CASRAI’s working paper and discussion paper dictionary entries for the fuller distinction from preprints.
Why does the working-paper number matter for citation?
Because the working-paper version is often what other researchers read and cite first—sometimes years before journal publication—the series and number (e.g., “NBER Working Paper No. 12345”) function as the de facto version of record during that interval. Citing the correct version (working paper vs. eventual journal article) matters for accurately representing what informed a given piece of downstream work.
Can working papers be revised after release?
Yes. Unlike a submitted journal manuscript, a working paper is commonly updated in place under its existing series number as the author incorporates feedback, which is one of the clearest procedural differences from journal submission, where a fixed version enters formal review.
This guide focuses on submission and formatting conventions for economics working-paper series generally; researchers should always confirm current requirements directly with NBER, IZA, or CEPR, since affiliation and formatting requirements are set and updated by each institution.
