Proposition 2, p. 12 — every innovation equilibrium is a duopoly; T*(β) is an equilibrium, boundary if k2 < k̂2(k1|β), interior otherwise
ProvedCrossLicensing.Innovation.proposition_2Assume and . In the innovation game, manufacturer chooses and earns , where is its Cournot profit at the supplier's optimal wholesale price (the least maximizer of over ). Let if and otherwise, with the closed forms of the proof. Then:
- every pure Nash equilibrium of the innovation game is a downstream duopoly (both manufacturers sell a positive quantity);
- , the profile is a pure Nash equilibrium, and it is a duopoly;
- (boundary duopoly) if , then , and the superior manufacturer invests exactly up to the duopoly boundary:
- (interior duopoly) if , then and both first-order conditions hold there:
with Lemma 1's duopoly gross profits.
Proposition 2 says that a supplier committing to cross-licensing never ends up with a single seller: the superior manufacturer either stops at the boundary where the supplier still prefers a duopoly, or innovates at an unconstrained interior level. Proposition 3 is its instance .
Formalization Note The equilibrium is not unique in general (for some parameters a mirrored boundary profile with on its own boundary is also an equilibrium), so the statement asserts that is an equilibrium and that every equilibrium is a duopoly, not that is the only one. The first-order conditions are taken on Lemma 1's duopoly polynomials, because the game payoff of jumps at . The paper's "or equivalently " is not formalized; "this region shrinks to empty as increases" is the milestone on 's monotonicity.
Retired 2026-10-27: a Lean-vs-paper fidelity audit found this statement weaker than, or narrower than, the cited result. A corrected statement will be resubmitted.
import Mathlib import Definitions.Def_CrossLicensing_Innovation_Model import Definitions.Def_CrossLicensing_Innovation_ClosedForms
namespace CrossLicensing.Innovation
theorem proposition_2 (β k1 k2 : ℝ) (h : Standing β k1 k2) :
(∀ T1 T2 : NNReal, IsInnovationEquilibrium β k1 k2 T1 T2 → IsDuopoly β T1 T2) ∧
(∃ T1 T2 : NNReal, (T1 : ℝ) = Tstar1 β k1 k2 ∧ (T2 : ℝ) = Tstar2 β k1 k2 ∧
IsInnovationEquilibrium β k1 k2 T1 T2) ∧
IsDuopoly β (Tstar1 β k1 k2) (Tstar2 β k1 k2) ∧
(k2 < khat2 β k1 →
β < Real.sqrt 3 - 1 ∧ Tstar1 β k1 k2 = TB1 β k1 ∧ Tstar2 β k1 k2 = TB2 β k1 ∧
Tstar2 β k1 k2 = That2 β (Tstar1 β k1 k2)) ∧
(khat2 β k1 ≤ k2 →
Tstar1 β k1 k2 = TI1 β k1 k2 ∧ Tstar2 β k1 k2 = TI2 β k1 k2 ∧
HasDerivAt (fun x => duoGross1 β x (Tstar2 β k1 k2) - k1 * x ^ 2) 0 (Tstar1 β k1 k2) ∧
HasDerivAt (fun x => duoGross2 β (Tstar1 β k1 k2) x - k2 * x ^ 2) 0 (Tstar2 β k1 k2)) := by sorry
end CrossLicensing.Innovation
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.