Proof of Proposition 7, p. 37 — at stage 3 with q_R > 0, an optimal (u_M, q_M) never has u_M = u_R and q_M > 0
ProvedQualityEncroach.NoCommit.stage3_no_uniformFix , , a wholesale price , a retailer quality and a retailer order . Suppose with , maximizes the manufacturer's stage-3 profit
over all , , where is the direct-channel market-clearing price of the two-quality model. Then it is impossible that both and :
In words: once the retailer has placed a positive order, the manufacturer never sells directly a product of the same quality as the retailer's. This is the contradiction that concludes the proof of Proposition 7, stated at every stage-3 history rather than only on the equilibrium path.
Formalization Note. The hypothesis is needed: the paper's contradiction comes from the strict signs printed in (31)–(32); with the weak signs that optimality gives it reads , which contradicts but not . At the manufacturer is a single-product monopolist and is optimal when is her monopoly quality.
import Mathlib import Definitions.Def_QualityEncroach_NoCommit_Game
namespace QualityEncroach.NoCommit
/-- Proof of Proposition 7, p. 37: at a stage-3 history with `u_R > 0` and `q_R > 0`, no
optimal choice `(u_M, q_M)` of the manufacturer (over `u_M > 0`, `q_M ≥ 0`) sells directly
(`q_M > 0`) at the retailer's quality `u_M = u_R`. -/
theorem stage3_no_uniform (k c : ℝ) (hk : 0 < k) (hc : 0 ≤ c) (w uR qR uM qM : ℝ)
(huR : 0 < uR) (hqR : 0 < qR) (huM : 0 < uM) (hqM : 0 ≤ qM)
(hopt : ∀ uM' qM' : ℝ, 0 < uM' → 0 ≤ qM' →
mfrPayoff k c ⟨w, uR, qR, uM', qM'⟩ ≤ mfrPayoff k c ⟨w, uR, qR, uM, qM⟩) :
¬ (uM = uR ∧ 0 < qM) := by sorry
end QualityEncroach.NoCommit
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.