Ep77 The Academic Journal System is Broken, Here’s How to Fix It

This episode is a discussion between Jules Van Binsbergen and Jonathan Burke about the state of academic publishing, especially in economics and finance, and about a new initiative they are launching called Informed Discourse. The core thesis is that the journal system is no longer well matched to modern distribution technology: when papers were scarce and expensive to circulate, centralized journal gatekeeping made sense, but now the internet has created an overabundance of papers and a severe curation problem. Their view is not that peer review is useless, but that the system has become inefficient, slow, and overly dependent on a few editors and referees whose incentives are misaligned with actually improving research. They spend much of the discussion on incentives. …

1. The speakers think the journal system is broken primarily because it wastes effort and delays knowledge transmission, not simply because it occasionally publishes bad papers.
2. They argue that reviewer incentives are weak: unpaid, anonymous referees tend to optimize for pleasing editors rather than deeply improving papers.
3. Long referee cycles push authors to make papers longer and more cumbersome, which reduces readability and creates negative externalities for the profession.
4. The internet solved distribution, but academia never built a good real-time curation layer to replace the old scarcity-based journal model.
5. Their proposed answer is an expert-only, non-anonymous, reputation-based platform that curates research by topic and relevance.
6. They see crowdsourced expert judgment as better than giving a few editors outsized control over publication and attention.
7. The project is explicitly framed as a service to the profession rather than a commercial or ideological venture.

• The claim that peer review quality is broadly as flawed now as in the past is asserted more than demonstrated.
• The idea that non-anonymous expert crowds will reliably produce better curation than journals is plausible but unproven in the transcript.
• Their skepticism toward editors may understate the value of centralized quality control in preventing low-quality or misleading work from spreading.
• The discussion assumes experts will participate enough to make the system useful; the cold-start problem is acknowledged but not resolved.
• They argue longer papers are mostly a referee-induced distortion, but no direct evidence is provided that length is net harmful rather than sometimes necessary.