Proposition 3(1),(3), p. 15 — the budget binds; patient-only subsidy if p ≥ 2/(5 − αmax), manufacturer-only if p ≤ 2(1 − αmax)/5
ProvedRareDisease.ExogOutcome.proposition_3Consider Setting (2B): the consortium fixes the unit price , the manufacturer is paid only when the drug is efficacious and best-responds to a subsidy policy by choosing its target efficacy , and the government chooses the policy to maximize patient welfare subject to spending at most the budget ; the R&D cost scale is . Let be the largest such that the policy is affordable at the manufacturer's best response. Then:
- (binding budget) at every optimal policy the spending at the manufacturer's best response equals ;
- (Proposition 3, part 1) if and , it is optimal to subsidize patients only: there is such that is an optimal policy, and every optimal policy equals ;
- (Proposition 3, part 3) if , it is optimal to subsidize the manufacturer only: is an optimal policy, every optimal policy equals it, and
The result says that under a consortium price with "No Cure No Pay" payment the government's whole budget goes to one side of the market when the price is high (patients) or low (the manufacturer).
Formalization Note Part 2 of the printed proposition (both subsidies when ) is not posed: it is false as printed (for , , the price lies in that range but the unique optimum is manufacturer-only). The printed closed form for in part 1 is not posed either, because it does not satisfy the binding constraint; is characterized by the binding clause instead. Part 1 adds the hypothesis : at the spending tends to as , so for larger budgets no patient-only policy exhausts the budget. The hypotheses , and are the paper's standing assumptions made explicit. "Optimal" quantifies over every policy and every best response; efficacies are not bounded by , as in the paper.
import Mathlib import Definitions.Def_RareDisease_ExogOutcome_Setting
namespace RareDisease.ExogOutcome
/-- Proposition 3, p. 15, parts 1 and 3 (with the binding budget constraint), Setting (2B).
`a` is `α_max`. Part 1 adds the disclosed hypothesis `B < p² / (2k)`; part 2 and the printed
closed form of part 1 are not posed (both are false as printed). -/
theorem proposition_3 (k B p : ℝ) (hk : 0 < k) (hB : 0 < B) (hp0 : 0 < p) (hp1 : p < 1)
(a : ℝ) (ha : IsAlphaMax k p B a) :
(∀ α β : ℝ, IsOptimal k p B α β → ∀ e ∈ BR k p α β, spend k p α β e = B) ∧
(2 / (5 - a) ≤ p → B < p ^ 2 / (2 * k) →
∃ β₁ : ℝ, IsOptimal k p B 0 β₁ ∧
∀ α β : ℝ, IsOptimal k p B α β → α = 0 ∧ β = β₁) ∧
(p ≤ 2 * (1 - a) / 5 →
IsOptimal k p B a 0 ∧
(∀ α β : ℝ, IsOptimal k p B α β → α = a ∧ β = 0) ∧
a = (8 * B * k + p ^ 4 - 2 * p ^ 3 + p ^ 2
- (p - p ^ 2) * Real.sqrt (16 * B * k + p ^ 4 - 2 * p ^ 3 + p ^ 2)) / (8 * B * k)) := by sorry
end RareDisease.ExogOutcome
Confirmed by the mission captain (proposal self-audit).
Confirmed by the moderator at approval.