nucleation

LIT-1326: Algorithmic monoculture and social welfare

1 min read

Properties5
StatusActive — read 2026-10-21 (NOTE-1033), from arXiv:2101.05853v2: the main text and Appendices A, B and E in full, Appendices C, D and F in statement; worth reading for the exact conditions under which a shared, more accurate screening algorithm is dominant for every firm and still lowers welfare without any shock (THEORY-835), and for the identity that makes independence valuable to a second mover. Read it knowing that the model has two firms hiring one candidate each, that the conditions hold for Mallows rankings always, for Gaussian and Laplacian random utility models with three candidates, and never for Plackett–Luce, and that the losses measured are small (about 4% in one instance).
Version2
Tagssocial-science, game-theory, anthology-candidate
Cited as “paper” byNOTE-1033 — Kleinberg and Raghavan — Algorithmic monoculture and social welfare
Cited as “source” byTHEORY-835 — When two firms each hire the best available candidate by a noisy ranking, and the ranking model makes an independent re-ranking worth more to the second mover than its own next choice and makes a weaker first mover better for it, then for every accuracy of the private rankings there is a more accurate shared algorithmic ranking that each firm strictly prefers to use whatever the other does, and with which both firms, and total welfare, do worse than if neither used it

Graph View

Kleinberg, J. & Raghavan, M. (2021). Algorithmic monoculture and social welfare. Proceedings of the National Academy of Sciences 118(22): e2018340118. doi:10.1073/pnas.2018340118. Preprint: arXiv:2101.05853 (14 January 2021).

Read 2026-10-21, in NOTE-1033: the main text and Appendices A, B and E in full, C, D and F in statement. The summary was checked against the reading.

Why this is here

A25: “Particularly relevant to common legibility regimes and loss of organizational diversity.” A26 §2 drew “a closely related formal warning” from it: decision-makers sharing one representation can do worse together. Bears on TERM-064. Not LIT-1103, which is Kleinberg’s small-world paper.

Backlinks