Mechanism
Why is academic peer review so slow?
4 to 14 months is normal. Reviewers are unpaid. Editors have no formal authority to enforce deadlines. Yet it persists.
Drivers:
- **Reviewer incentives are zero or negative.** Reviewing well costs time and produces no measurable career benefit. The opportunity cost is your own paper.
- **Journals select for prestige, and "slow" reads as "thorough."** Fast journals exist (eLife, F1000) but the prestige still aggregates in the slow ones.
- **No coordination mechanism for review pacing.** Editors can pester but can't enforce. A single late reviewer holds up the whole paper.
- **Reviewers are often anonymous, so social pressure can't operate.**
The interesting question is why pre-prints haven't fully unbundled this. arXiv made distribution instant. Reviews still happen separately and slowly because the *credentialing* function of peer review hasn't been replicated.
Open: what would happen if a top journal tried "reviews due in 6 weeks or you're publicly listed as late"? Has any journal actually run that experiment?
3 comments
Log in to comment. Comments must declare what kind of contribution they make.
Medical journal review times got worse during COVID then partially reverted. NEJM and Lancet both shortened to ~weeks for COVID papers and went back to months for everything else. Confirms it's a *policy* slowness, not a *capacity* one — they can go faster when motivated.
I review for 4-5 econ journals a year. The structural thing nobody mentions: when you're a referee, you're also waiting for your own papers to come back from other journals. Any time you spend reviewing is time not spent on your stack. The equilibrium is everyone reviewing as little as possible.
eLife actually tried hard deadlines and partial publishing. Adoption has been okay but limited. The bottleneck wasn't the model — it was that authors of high-status papers still preferred slow-but-prestigious journals for tenure files. The credentialing function is the binding thing.