Reinforcing Research Transparency at Management Science

In 2019, Management Science adopted a Code and Data Disclosure policy (https://pubsonline.informs.org/page/mnsc/code-and-data-disclosure-policy) with the goal of enabling the reproducibility of the research we publish. Since then, this policy has positioned the journal at the forefront of transparency in the management and decision sciences. Our initial efforts focused on algorithmic reproducibility and empirical replication. More recently, we have begun exploring the verification of source data in experimental studies, and we anticipate further developments in documentation standards and, eventually, even proof-checking as best practices continue to evolve.

The impact of this policy is real and measurable. A recent large-scale reproducibility audit—conducted with more than 700 reviewers and assessing nearly 500 articles published before and after the policy’s introduction—found that more than 95% of articles subject to the new policy could be computationally reproduced when data and technical access were available (Fišar et al. 2023). This is a dramatic improvement from the prepolicy period, when only 12% of articles voluntarily provided replication materials and fewer than half of those could be reproduced. A separate replication study of experimental operations management papers further underscored the effect of these policies on research confidence and robustness (Davis et al. 2023).

However, this progress has introduced new operational challenges. The volume and complexity of reproducibility checks have outgrown what can be managed using volunteer editorial resources alone. Today, we face a growing backlog of papers accepted in principle but awaiting verification of their data packages before they can appear in “Articles in Advance,” our preprint publication queue for accepted articles. This bottleneck reflects the limits of what we can manage without additional infrastructure or support. (In rare cases, unresolvable issues discovered during verification could affect the final publication decision.) To scale this effort responsibly, we have partnered with the Certification Agency for Scientific Code and Data (cascad, https://www.cascad.tech/), a nonprofit certification agency founded by academics and supported by the French National Science Foundation (CNRS) and a network of French research institutions. Although mission aligned, this collaboration necessarily involves new and recurring costs. Appropriate safeguards are in place to maintain confidentiality, prevent conflicts of interest, and protect our authors’ intellectual property consistent with journal policies and INFORMS-wide ethical standards.

To sustain and expand the reproducibility program, Management Science will begin assessing a $79 submission fee for all original submissions, effective August 1, 2025. This fee is modest by industry standards and was selected based on financial modeling and feedback from a broad community survey conducted in December 2024, which also helped shape both the amount and the waiver policy. We are grateful to everyone who provided feedback.

The submission fee will directly fund the journal’s reproducibility infrastructure and external verification support and is also intended to support expanded research dissemination. Its implementation reflects our longstanding commitment to transparency and research integrity—not only in what the funds support but in how the process is managed, communicated, and monitored. If excess revenue is generated, remaining funds will be reinvested into INFORMS’ publications program, with a focus on strengthening the journal ecosystem.

Because submission fees remain relatively uncommon in the social sciences and management disciplines, we have taken particular care to ensure ours is both practical and accessible—with broad waiver eligibility, a low base fee, and an implementation approach rooted in trust and inclusion.

At present, no similar fee is planned for other INFORMS journals. Although several INFORMS journals are expanding their reproducibility efforts, the scale and complexity of implementation at Management Science remains unique at this time—due in part to our high submission volume and methodological breadth.

We are committed to ensuring that the fee does not become a barrier to participation. The fee will be waived for all INFORMS members, authors whose primary affiliation is in a low- or lower-middle-income country (as designated by the World Bank), and any author who requests a waiver due to financial hardship. Waivers will be granted on an honor system—no explanation required.

Some in our community have asked why we chose a submission fee rather than a publication (acceptance) fee. In theory, an acceptance fee might seem more equitable, but it would present two practical challenges. First, it would delay cost recovery by years because many accepted papers already in the verification backlog would incur costs without contributing to cost recovery. Second, it would require a significantly higher fee per accepted paper, raising additional concerns about access and equity. The submission fee model allows us to distribute costs more broadly, offer generous waivers, and keep the fee as low and sustainable as possible.

We have taken care to implement this policy in a way that reflects our values and serves the community. The ScholarOne submission system will include a simple declaration form at the point of submission, with waiver options presented upfront. An author’s ability to pay will have no bearing on the editorial review or decision process—and will in fact be completely invisible to everyone within the editorial decision-making hierarchy. Additional guidance, including a public FAQ, is available on our website (https://pubsonline.informs.org/journal/mnsc). Although we expect most of the funds to support Management Science directly, a portion may be used to sustain and strengthen INFORMS’ broader publishing infrastructure—which ultimately benefits our authors and community.

We are deeply grateful for the support of our editorial board, INFORMS leadership, and our authors, readers, and reviewers. With this policy, we reaffirm our shared commitment to integrity, equity, and excellence in scholarly publishing.

References

  • Davis AM, Flicker B, Hyndman K, Katok E, Keppler S, Leider S, Long X, et al. (2023) A replication study of operations management experiments in Management Science. Management Sci. 69(9):4977–4991.LinkGoogle Scholar
  • Fišar M, Greiner B, Huber C, Katok E, Ozkes AI (2023) Reproducibility in Management Science. Management Sci. 70(3):1343–1356.LinkGoogle Scholar