p. 159 — if (1, 1) maximizes u₁u₂ on a convex S, no point of S has u₁ + u₂ > 2
ProvedNashBargainingProblem.Axiomatic.sum_le_two_of_one_one_maxbargainingconvexitynash-bargaining-solutionp2o-batch-pfp2bp2o-gran-per-chapterp2o-plan-paperp2o-v1
Let be convex, let , and suppose is a point of maximum of over in the closed first quadrant: for every with . Then
In Nash's proof this places the whole set of alternatives on one side of the line , which touches the hyperbola at (Figure 1).
Formalization Note The conclusion holds for every point of , not only for those in the first quadrant, as the paper's "for no points of the set" says. The hypothesis is the weak maximum (), which is what the paper's argument uses; the strict maximizer of the previous steps satisfies it.
Preamble
import Mathlib
Formal statement
namespace NashBargainingProblem.Axiomatic
theorem sum_le_two_of_one_one_max (S : Set (ℝ × ℝ))
(hS_convex : Convex ℝ S) (h11 : ((1 : ℝ), (1 : ℝ)) ∈ S)
(hmax : ∀ s ∈ S, 0 ≤ s.1 → 0 ≤ s.2 → s.1 * s.2 ≤ 1) :
∀ u ∈ S, u.1 + u.2 ≤ 2 := by sorry
end NashBargainingProblem.Axiomatic
Source
Nash, The Bargaining Problem, Econometrica 18 (1950), p. 159, Two Person Theory, proof of the assertion ("For no points of the set will u1 + u2 > 2, now, since …"), Figure 1 (p. 160)
Human review
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.