“Certified random” author order is a specific, named mechanism from economics: co-authors use the American Economic Association’s (AEA) online Author Randomization Tool to generate a genuinely random byline order, sign a consent form agreeing to publish in that order, and mark the result with the symbol “ⓡ” between author names so readers know the order carries no meaning about relative contribution. It is not a generic description of “randomizing author order” — it is this particular tool, symbol, and paper-trail, proposed by Debraj Ray and Arthur Robson in a 2018 American Economic Review paper and now offered as an official AEA service.
Where the idea came from: the Ray-Robson paper
The mechanism traces to a specific, citable source: Debraj Ray and Arthur Robson, “Certified Random: A New Order for Coauthorship,” American Economic Review 108, no. 2 (2018): 489–520. The paper’s starting problem is the one economics has long lived with: because alphabetical-by-surname order is the field’s dominant convention (see how the same alphabetical norm developed independently in mathematics), first authorship in a two-author economics paper correlates with an author whose surname happens to fall early in the alphabet — not with who did more of the work. Being listed and cited first (“Smith et al.” in short form) carries a real citation and career advantage that has nothing to do with contribution.
Ray and Robson’s proposed fix does not abandon the alphabet in favor of a US-biomedicine-style position-coded order (first-author-did-most, last-author-is-senior — see how that convention works in the life sciences). Instead it keeps the flat, unranked byline that economics is used to, but replaces the alphabetical tiebreaker with a genuinely random one, and makes that randomness publicly verifiable with a symbol. Their paper argues this approach is robust to gradual adoption (a “certified random” paper isn’t disadvantaged if surrounded by alphabetical ones), distributes the first-position advantage evenly over time, and still lets co-authors signal a real contribution asymmetry by choosing NOT to randomize when contributions are genuinely unequal.
How the AEA’s Author Randomization Tool works
The AEA now runs this as a live, official service at aeaweb.org/journals/random-author-order, built directly on the Ray-Robson mechanism. In practice, the process has three parts:
- Generate the order. Co-authors use the AEA’s tool to produce a random ordering of their names.
- Archive the result. The AEA states the outcome “will be archived so that you and your coauthors have a record of the randomization and its results” — creating a retrievable record that the order was genuinely randomized, not a description the authors simply asserted after the fact.
- Sign a consent form. Co-authors submit a signed consent form confirming they agree to publish under the random order the tool produced. This is what makes the process “certified” rather than just “random”: the paper trail (archived randomization plus signed consent) is what a reader, editor, or tenure committee could in principle verify, rather than taking the byline order on faith.
The tool is offered as a service for AEA-published journals (including the American Economic Review). It is the best-documented, official implementation of the Ray-Robson mechanism; the search here did not find confirmed adoption of an equivalent official tool by other major economics journals (Econometrica, the Quarterly Journal of Economics, the Journal of Political Economy), though the ⓡ symbol itself, since it originates from a published methodology rather than a proprietary AEA mark, is understood across economics as signaling the same thing wherever an author byline uses it.
The ⓡ symbol: what it means and how it’s used
The symbol is a lowercase letter “r” inside a circle: ⓡ. It is easy to confuse visually with the registered-trademark mark “®” (a different, unrelated character) — a mix-up worth watching for if you’re copying it from a PDF, since some PDF text-extraction tools render it incorrectly as “®”. In a byline, ⓡ replaces the word “and” between the names it separates, for example:
Debraj Ray ⓡ Arthur Robson
According to the AEA, the symbol “signals to the reader that the authors’ names have been randomized using a uniform distribution” — in other words, every possible ordering of the co-authors’ names had an equal chance of being the one published. The convention extends to spoken and in-text citation too: the AEA suggests using the phonetic sound of the letter “r” in place of “and” when citing the paper aloud, and papers are commonly cited in text as, for example, “Ray ⓡ Robson (2018)” rather than “Ray and Robson (2018).”
Because the symbol only means “this specific order was randomized and certified,” it should not be used informally to mean “these authors are listed in no particular order” without going through the actual tool-and-consent process — doing so would misrepresent the byline the same way an uncertified claim of alphabetical-only ordering would.
The consent-form requirement
The signed consent step exists for the same reason ICMJE requires every author to approve the final version of a manuscript before publication (see ICMJE’s four authorship criteria): author order, even when randomized, is a substantive claim about the paper that every named author needs to actually agree to, not something one co-author can impose on the others unilaterally. Requiring signed consent before the randomized order is finalized and published prevents a scenario where the tool generates an order one co-author is unhappy with and the paper proceeds anyway. It also creates a documented record that all co-authors specifically agreed to random (rather than contribution-based or alphabetical) ordering, which matters if author order is later questioned by an institution, funder, or the authors themselves.
Why economics adopted this instead of a position-coded system
Economics could, in principle, have moved toward the position-coded convention biomedicine uses, where first author signals primary intellectual and hands-on contribution and last author signals the senior/supervising role. Ray and Robson’s paper argues that a randomization-plus-symbol approach fits economics better for a specific reason: it preserves the field’s existing norm of a flat, unranked byline (which many economics collaborations, especially long-running ones between equal co-authors, prefer) while still solving the specific fairness problem — that “first” in an alphabetical list has nothing to do with contribution — without requiring every project to negotiate and defend a contribution ranking. Co-authors who genuinely do have unequal contributions remain free to signal that separately (for example, by not randomizing, or through other disclosure), so certified random order is an opt-in fix for the specific case where contributions really are comparable and alphabetical order was just an accident of surname.
How certified random order compares to other author-order conventions
Certified random order is one of several distinct solutions different fields have developed for the same underlying problem — that a flat author list conveys no information about who did what, and whatever ordering rule fills that gap ends up conferring an advantage that may have nothing to do with actual contribution:
- Alphabetical order (mathematics, and historically economics) — simple and free of negotiation, but confers a durable, contribution-blind advantage on early-alphabet surnames. See Mathematics’ Alphabetical Author Order: Origins and How It’s Changing.
- Position-coded order (biomedicine and most life sciences) — first author did the most hands-on work, last author supervised; middle positions are comparatively ambiguous. See Middle Author vs. Last Author and Co-First Authorship for how ties are handled.
- Certified random order (economics, via the AEA tool) — keeps a flat, unranked list but removes the alphabetical advantage through verified randomization.
- Collaboration-defined order (high-energy physics, astronomy megacollaborations) — with author lists sometimes running into the thousands, order is set by the collaboration’s own membership rules rather than by paper-by-paper negotiation. See Authorship in High-Energy-Physics Megacollaborations.
- Regional variations on Western defaults — author-order signaling conventions also differ by country in ways that don’t map cleanly onto any of the above; see Author Order in China, Japan, and Korea.
For the full discipline-by-discipline picture, including where certified random order sits relative to CRediT contribution statements (which describe what each author did, independent of byline order), see the overview guide How Authorship Order Is Decided: Conventions Across Disciplines.
Frequently asked questions
Is the ⓡ symbol the same as the ® registered-trademark symbol?
No. They are different Unicode characters that look similar. ⓡ is a lowercase “r” inside a circle, used specifically to signal certified-random author order in economics. ® marks a registered trademark and has no connection to authorship. The two are sometimes confused when text is copied from a PDF, since some extraction tools render ⓡ incorrectly as ®.
Do co-authors have to use the AEA’s official tool, or can they just randomize order themselves and add the symbol?
The AEA’s own process requires generating the order through its tool, archiving the result, and signing a consent form — that combination is what makes the order “certified,” not just randomly chosen. Using the ⓡ symbol without going through an actual randomization-and-consent process would misrepresent the byline.
Does certified random order replace a CRediT contribution statement?
No, the two do different jobs. Certified random order concerns only the sequence names appear in the byline; a CRediT contribution statement separately describes which specific tasks (conceptualization, formal analysis, writing, and so on) each author performed. A paper can use certified random order for its byline and still publish a full CRediT statement alongside it.
Can co-authors choose not to randomize if contributions are unequal?
Yes — certified random order is an opt-in mechanism intended for cases where contributions are genuinely comparable and an alphabetical tiebreaker would otherwise be arbitrary. Ray and Robson’s original proposal explicitly treats the choice not to randomize as itself a signal that one author’s contribution was substantially larger.
Is certified random order used outside economics?
The AEA’s official tool is specific to AEA-published journals, and the mechanism originated in and remains most closely associated with economics. It has not become a standard convention in most other fields, which continue to use alphabetical order, position-coded order, or collaboration-defined order depending on discipline (see the comparisons above).
