Submission quotas as self-screening when manuscript production is cheap
summary
The gist
Generative AI enables authors to produce more manuscripts at lower cost.
In short
The paper compares submission quotas and common fees for authors submitting manuscripts, especially with cheap generative AI tools. It derives a mathematical condition showing when quotas provide a weakly higher mean reviewed value than fees, given an equal expected review load. The findings suggest quotas are superior when manuscript values are bounded and candidate manuscripts are abundant.
Key concepts
- Submission Quotas
- Quotas transform submission into a revealed-priority decision where authors build a portfolio of their highest-ranked manuscripts. This mechanism forces authors to prioritize their work, potentially leading to better overall quality submissions compared to simple fees.
- Common Fee Mechanism
- A common fee charges an author for every manuscript submitted. This mechanism allows for more flexible submission thresholds but can lead to heterogeneous submission rates based on an author's baseline portfolio quality.
- Mean Reviewed Value
- This metric represents the average quality of manuscripts reviewed by expert editors and referees. The paper derives a condition where quotas yield a weakly higher mean reviewed value than fees when expected review loads are equal.
- Portfolio Construction
- Authors construct a portfolio of their highest-ranked manuscripts to satisfy quota requirements. This process is central to the quota mechanism, as it dictates which submissions are prioritized based on the author's perceived quality ranking.
Terminology used across episodes
This episode discusses
- Submission quotas as self-screening when manuscript production is cheap · Paper Radio
- Submission Responsibility Matters: Role-Aware Submission Quotas under Coauthorship
- How Many Submissions May an Author Make? A Harmonic Quota for Submissions under Coauthorship
The paper
Submission quotas as self-screening when manuscript production is cheap · Read on arXiv
Steve Lawford
ENAC (University of Toulouse)
Transcript
Introduction to the show: ident: Economics. The day in economics: theory, mechanisms and the evidence beneath them.
Ruth: We're looking at the paper: "Submission quotas as self-screening when manuscript production is cheap".
Warren: Here's the gist of the paper. Generative AI enables authors to produce more manuscripts at lower cost.
Ruth: The paper derives a condition under which submission quotas yield weakly higher mean reviewed value than fees, given an equal expected review load.
Warren: The core mechanism of the quota versus fees A common fee charges an author for every manuscript they submit.
Ruth: Quotas, however, transform submission into a revealed-priority decision by constructing a portfolio of highest-ranked manuscripts.
Paper summary: Ruth: So, this paper is called "Submission quotas as self-screening when manuscript production is cheap," and it’s looking at how journals are responding to the rise of generative AI. The core idea here is that since authors can now make way more manuscripts cheaply, the value of actually submitting something to get expert review becomes really important.
Warren: Right, so they’re comparing two main ways journals are trying to manage this: using submission quotas or charging a fee for every single manuscript an author sends in. It seems like the paper is investigating under what conditions one method is better than the other when you have to balance things out, like everyone gets the same expected amount of review capacity.
Iris: That sounds like they’re setting up a comparison between two different incentive structures, which is interesting because it moves away from just looking at volume and starts looking at how authors actually decide what to submit based on their own internal ranking of work.
Ruth: Exactly, and the paper claims that non-transferable author quotas turn submission into a revealed priority decision. It means an author has to choose their highest-ranked work because they can't just pass the submission right on to someone else, which creates an opportunity cost for choosing one manuscript over another <ref:2610.04688#pg2>.
Warren: And they argue that this portfolio construction is what makes the quota mechanism different from a simple flat fee structure. The fee just charges for every submission, whereas the quota forces authors to curate their best set of candidates first, which is a different kind of selection <ref:2610.04688#pg1>.
Iris: So they are trying to identify the precise mathematical condition where that portfolio approach actually beats the fee approach when we assume everyone has the same expected load on reviewers. It’s about finding when you get more value from a curated choice versus just paying for every attempt <ref:2610.04688#pg3>.
Ruth: That condition they derive, which is written as sigma two d plus two gamma greater than or equal to tau of M and K, seems to be the key result. It basically tells us when the dispersion in those submission cutoffs is bigger than the dispersion in an author's baseline portfolio quality <ref:2610.04688#pg3>.
Warren: And that condition also has an equivalent way of being stated, which is sigma two c minus sigma two theta greater than or equal to tau of M and K. That relates the fee structure’s dispersion to how good an author's baseline work is compared to the threshold they set <ref:2610.04688#pg3>.
Paper summary: Iris: I wonder what this means for identification, because it relies on knowing that authors have private information about their manuscript quality, which is a huge assumption in these models. What happens if that private information isn't perfectly captured?
Ruth: That’s a fair point, Iris. The paper acknowledges that the comparison between quota and fee depends on the dispersion of fee-induced selectivity across authors and how that relates to baseline portfolio quality <ref:2610.04688#pg3>. It's not just an abstract mathematical thing; it ties back to how different types of authors react to these policies.
Warren: The paper does show some real implications for policy design, specifically that when project values are bounded and the candidate manuscripts are abundant, an equal-load fee can actually price out some authors completely <ref:2610.04688#pg1>. That’s a big warning for journals trying to manage submission volume through fees alone.
Iris: So it suggests that fees might be too blunt an instrument because they could systematically exclude certain author types entirely if their baseline quality doesn't meet the fee threshold <ref:2610.04688#pg1>. It’s not just about filtering out some submissions; it’s about filtering out entire authors.
Ruth: Precisely, and this points toward the idea that quotas might be a better way to reserve review capacity across the active pool of authors, rather than letting fees price out people who are still contributing valuable work <ref:2610.04688#pg3>.
Warren: The paper also suggests that quotas should probably be applied to first-round submissions rather than just acceptances because that acceptance-based rule doesn't require the author to prioritize anything before they even submit <ref:2610.04688#pg3>. It seems like a quota forces that prioritization upfront.
Iris: And the authors also suggest a combination of fee and quota might be beneficial, where the fee can control how intense an author is submitting overall while the quota prevents any single author from flooding the system with too many high-quality options <ref:2610.04688#pg3>.
Ruth: That’s a practical implication for policy design, showing that maybe neither pure method is perfect on its own, and they also mention that the accounting unit matters in this comparison <ref:2610.04688#pg3>.
Warren: And remember, the paper itself notes a limitation: the model assumes candidate sets are already given to us, even though in reality, submission rules might actually affect what research effort an author puts into their projects <ref:2610.04688#pg3>. It’s a simplification of how things actually work.
Paper summary: Iris: So the paper lays out this condition for when quotas yield higher mean reviewed value than fees, given that equal expected review load, but the authors themselves have to be careful about assuming those candidate sets are fixed <ref:2610.04688#pg3>. It’s a specific scenario where their comparison holds true.
Ruth: Moving toward the conclusion of "Submission quotas as self-screening when manuscript production is cheap," the authors are essentially saying that for certain scenarios—when project values are bounded and there are plenty of candidate manuscripts—an equal-load fee has the potential to price out some authors entirely <ref:2610.04688#pg1>.
Warren: The big picture here is that quotas create an opportunity cost, forcing authors to prioritize their work, which seems like a way to reserve review capacity for those who have already demonstrated a strong commitment to quality <ref:2610.04688#pg2>.
Iris: It’s about how the mechanism of selection changes the outcome when you introduce this new context of cheap manuscript production, showing that the structure of the submission process really dictates what kind of authors are incentivized to submit and what reviewers end up seeing <ref:2610.04688#pg3>.
Ruth: So, in simple terms, they’re saying that when you have a lot of cheap AI-generated work, journals need a way to stop authors from just flooding the system with low-effort submissions and instead encourage them to curate their best work through a quota system <ref:2610.04688#pg1>.
Warren: And they show that the mathematical condition for quotas being better than fees depends on the spread between how selective different authors are versus how good their baseline work is, which is where the real friction in this comparison lies <ref:2610.04688#pg3>.
Iris: It’s a nuanced result, showing that while quotas might be better for reserving capacity among active authors, the fee structure has its own way of filtering out author types depending on those dispersion parameters <ref:2610.04688#pg1>.
Ruth: That’s the main thing to remember about this paper: it gives a specific mathematical benchmark for when you should prefer one incentive structure over the other in this new publishing environment <ref:2610.04688#pg3>.
Warren: And that condition, sigma two d plus two gamma greater than or equal to tau of M and K, is the formal way to put it for when quotas win out over fees under equal review load <ref:2610.04688#pg3>.
Iris: So for anyone listening who just wants the practical implication, it’s that journals should consider how these structural choices affect the actual quality of reviews they get versus just looking at submission counts <ref:2610.04688#pg1>.
Ruth: And that’s what we have for this discussion on "Submission quotas as self-screening when manuscript production is cheap."
Conclusion: Ruth: So, we've been looking at how journals are trying to manage submissions when AI lets authors produce tons of work cheaply, and this paper is titled "Submission quotas as self-screening when manuscript production is cheap."
Warren: It boils down to comparing two ways they try to control the influx of material: charging a fee for every submission versus using submission quotas.
Iris: The main finding here is that under certain conditions, the quota system actually yields a higher mean reviewed value than the flat fee structure, assuming everyone gets an equal expected review load.
Ruth: That's what they're trying to prove mathematically. They derive a specific condition—sigma two d plus two gamma greater than or equal to tau of M and K—that tells you when the quota wins out over the fee approach.
Warren: And they show that when the manuscript values are bounded and there are lots of candidate manuscripts available, an equal-load fee can actually price out some authors completely. That's a strong result for how journals might be trying to filter things out.
Iris: The authors suggest quotas create an opportunity cost by forcing authors to prioritize their work before submitting anything. That means they are essentially forcing self-screening before the submission even happens.
Ruth: And the paper points out that this prioritization is key. It's not just about volume; it’s about how an author chooses which of their high-ranked manuscripts they actually put forward for review.
Warren: So, if you only listen to this, think about how a journal might stop authors from just flooding the system with low-effort submissions and instead encourage them to curate their best work through a quota system.
Iris: It’s important to remember that this whole comparison depends on assumptions—specifically, that the candidate manuscript sets are given upfront. That’s a limitation they flag in the paper.
Ruth: Right, so we've seen the math showing when quotas beat fees and what it means for authors having to think about their portfolio before they submit. Next up, we look at how this structure actually plays out in real policy design and what those implications are for academic publishing right now.
More episodes
- Daily Summary for 2026-11-19
- 2608.21274-Recommendation Quality and the Concentration of Consumption: Experimental Evidence from Netflix
- 2408.02757-A nonparametric test for diurnal variation in spot correlation processes
- 2306.16393-High-Dimensional Canonical Correlation Analysis
- 2605.17299-Geometric Brownian motion with intermittent entries and exits
- 2607.04753-Fooling Yourself: how narratives shape beliefs
- 2408.14671-Inference on Treatment Effects with Repeatedly Mismeasured High-Dimensional Controls
- 2201.06898-Difference-in-Differences Estimators for Treatments Continuously Distributed at Every Period
- 2501.15307-Influence Functions: A Unified Differentiation-Based Framework
- The day in conversation for 2026-11-19