Proposition 1, p. 10 — under strong network effects the stage-2 equilibria are all-buy, none-buy, or both
ProvedNetworkQuality.PotentialMax.proposition_1In the network market under the standing assumptions, fix a quality with strong network effects, i.e. . Then for every price the (pure-strategy) Nash equilibria of the stage-2 subgame are exactly as follows:
- if , the only equilibrium is (all consumers buy);
- if , the only equilibrium is (no consumer buys);
- if , the equilibria are exactly and .
Under strong network effects consumers always coordinate on a common action, and for a range of prices both coordinated outcomes are equilibria; this multiplicity is what makes equilibrium selection matter for the firm.
Formalization Note "Equilibrium" is read as pure-strategy Nash equilibrium of stage 2, identified with the set of buyers, and each case is stated as a characterization of the set of equilibria, which is what the paper's proof establishes. and are the Lean terms M.v ⟨0, _⟩ and M.v ⟨n - 1, _⟩. The hypothesis is the paper's domain of quality.
import Mathlib import Definitions.Def_NetworkQuality_PotentialMax_Setting
namespace NetworkQuality.PotentialMax
open Market
theorem proposition_1 {n : ℕ} (M : Market n) (hM : M.Standing) (q : ℝ) (hq : 0 ≤ q)
(hs : M.StrongNE q) (p : ℝ) :
(p < M.v ⟨0, hM.1⟩ * M.f 1 q →
∀ B : Finset (Fin n), M.IsNash B p q ↔ B = Finset.univ) ∧
(M.v ⟨n - 1, Nat.sub_lt hM.1 Nat.one_pos⟩ * M.f n q < p →
∀ B : Finset (Fin n), M.IsNash B p q ↔ B = ∅) ∧
(M.v ⟨0, hM.1⟩ * M.f 1 q ≤ p → p ≤ M.v ⟨n - 1, Nat.sub_lt hM.1 Nat.one_pos⟩ * M.f n q →
∀ B : Finset (Fin n), M.IsNash B p q ↔ (B = Finset.univ ∨ B = ∅)) := by sorry
end NetworkQuality.PotentialMax
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.