Disproportionate division requires 2n−2 cuts

Informal Conjecture (f(n) = 2n−2) · arXiv:1909.07141

arXiv Informal medium confidence— first stated 2019-09-16

Authors: Logan Crew, Bhargav Narayanan, Sophie Spirkl

Status open high confidence

The conjecture that $f(n) = 2n-2$ for all $n \geq 2$ remains open as of May 2026. The source paper (published in Bulletin of the London Mathematical Society, 2020) establishes $f(n) \leq 3n-4$, improving the earlier O(n \log n) bound, while the matching lower bound $f(n) \geq 2n-2$ comes from a construction of Segal-Halevi. No follow-up paper closing the gap between $2n-2$ and $3n-4$ has been found in the indexed literature. Narayanan's publication list through 2026 contains no further work on this problem.

Reviewer notes. The arXiv abstract mentions a topological conjecture (unproven) that would imply 2n-2 cuts suffice in general, which together with the Segal-Halevi lower bound would settle f(n)=2n-2. No paper resolving this topological conjecture or otherwise closing the gap was found after five web calls.

Auto-reviewed 2026-05-15 with claude-sonnet-4-6 (web search enabled).

Informal. We suspect that the above construction reflects the truth and that $f(n) = 2n - 2$ for all $n \geq 2$.

Context

The authors give a construction (attributed to Segal-Halevi [6]) showing $f(n) \geq 2n-2$ for all $n \geq 2$, where $f(n)$ is the maximum number of cuts needed for disproportionate division with $n$ agents. Theorem 1.1 establishes $f(n) \leq 3n-4$, leaving a gap. The authors believe the lower bound is tight and that the truth is $f(n) = 2n-2$.

Source paper

Disproportionate division
Logan Crew, Bhargav Narayanan, Sophie Spirkl · 2019-09-16
https://arxiv.org/abs/1909.07141 PDF source