Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 3(a) — general development policy

Proved
WangGreenTech.Regulation.proposition_3a

by Shuze Chen · Oct 29, 2026 · Mathlib 0df444a (Lean v4.33.1)

global-gamesoperations-researchp2o-gran-per-chapterp2o-plan-paperp2o-v1regulation

Let p=(ϵ,b,q,r,f1,fL,fH,mL,mH)p=(\epsilon,b,q,r,f_1,f_L,f_H,m_L,m_H)p=(ϵ,b,q,r,f1​,fL​,fH​,mL​,mH​) satisfy ϵ>0\epsilon>0ϵ>0, b≥0b\ge0b≥0, 0≤q<10\le q<10≤q<1, r>0r>0r>0, f1,fL>0f_1,f_L>0f1​,fL​>0, fH≥fLf_H\ge f_LfH​≥fL​, and mL≤mH≤0m_L\le m_H\le0mL​≤mH​≤0. Signals and the common fundamental are real. Conditional on signal xxx, the specified posterior is uniform on [x−ϵ,x+ϵ][x-\epsilon,x+\epsilon][x−ϵ,x+ϵ]. Followers form a unit-mass population on [0,1)[0,1)[0,1); the leader has zero population mass. Both choose action 1 at indifference.

For C≥0C\ge0C≥0, define

B(ϵ,C,r)=inf⁡({2ϵ+Ctrt:0<t≤1}∪{2ϵ+C(1−sr)1−s:0≤s<1}).B(\epsilon,C,r)=\inf\left(\left\{\frac{2\epsilon+Ct^r}{t}:0<t\le1\right\}\cup\left\{\frac{2\epsilon+C(1-s^r)}{1-s}:0\le s<1\right\}\right).B(ϵ,C,r)=inf({t2ϵ+Ctr​:0<t≤1}∪{1−s2ϵ+C(1−sr)​:0≤s<1}).

Write b(2)=B(ϵ,(1−q)(fH−mL),r)b^{(2)}=B(\epsilon,(1-q)(f_H-m_L),r)b(2)=B(ϵ,(1−q)(fH​−mL​),r) and b(1)=B(ϵ,(1−q)(mH−mL),r)b^{(1)}=B(\epsilon,(1-q)(m_H-m_L),r)b(1)=B(ϵ,(1−q)(mH​−mL​),r).

The prescribed endogenous adoption threshold is t=b/2+fL−fH+rr+1(1−q)(fH−mL)t=b/2+f_L-f_H+\frac r{r+1}(1-q)(f_H-m_L)t=b/2+fL​−fH​+r+1r​(1−q)(fH​−mL​), with aggregate At(θ)=min⁡{1,max⁡{0,(θ+ϵ−t)/(2ϵ)}}A_t(\theta)=\min\{1,\max\{0,(\theta+\epsilon-t)/(2\epsilon)\}\}At​(θ)=min{1,max{0,(θ+ϵ−t)/(2ϵ)}}. Define

UD(λ,x)=12ϵ∫x−ϵx+ϵ[λ−f1+mL+θ−bAt(θ)+(q+(1−q)At(θ)r)(mH−mL)]dθ.U_D(\lambda,x)=\frac1{2\epsilon}\int_{x-\epsilon}^{x+\epsilon}\left[\lambda-f_1+m_L+\theta-bA_t(\theta)+(q+(1-q)A_t(\theta)^r)(m_H-m_L)\right]d\theta.UD​(λ,x)=2ϵ1​∫x−ϵx+ϵ​[λ−f1​+mL​+θ−bAt​(θ)+(q+(1−q)At​(θ)r)(mH​−mL​)]dθ.

The development set is Dλ={x:UD(λ,x)≥0}D_\lambda=\{x:U_D(\lambda,x)\ge0\}Dλ​={x:UD​(λ,x)≥0} and xaggr(1)(λ)=inf⁡Dλx_{\rm aggr}^{(1)}(\lambda)=\inf D_\lambdaxaggr(1)​(λ)=infDλ​.

Assume b≤b(2)b\le b^{(2)}b≤b(2). Then 2ϵ≤b(1)<b(2)2\epsilon\le b^{(1)}<b^{(2)}2ϵ≤b(1)<b(2); b(1)b^{(1)}b(1) is strictly increasing in ϵ>0\epsilon>0ϵ>0 and in mH−mL≥0m_H-m_L\ge0mH​−mL​≥0 with q,rq,rq,r fixed. For every λ≥0\lambda\ge0λ≥0, DλD_\lambdaDλ​ is nonempty and bounded below, and the threshold follower profile together with development on DλD_\lambdaDλ​ is a specified-belief equilibrium.

If b≤b(1)b\le b^{(1)}b≤b(1), then for every λ≥0\lambda\ge0λ≥0,

Dλ=[xaggr(1)(λ),∞).D_\lambda=[x_{\rm aggr}^{(1)}(\lambda),\infty).Dλ​=[xaggr(1)​(λ),∞).

If b>b(1)b>b^{(1)}b>b(1), there exist real λ‾aggr≤λ‾aggr\underline\lambda_{\rm aggr}\le\overline\lambda_{\rm aggr}λ​aggr​≤λaggr​, independent of λ\lambdaλ, such that for each λ≥0\lambda\ge0λ≥0 there are real x≤y≤zx\le y\le zx≤y≤z, with x=xaggr(1)(λ)x=x_{\rm aggr}^{(1)}(\lambda)x=xaggr(1)​(λ), satisfying

Dλ=[x,∞)if λ<λ‾aggr or λ>λ‾aggr,D_\lambda=[x,\infty)\quad\text{if }\lambda<\underline\lambda_{\rm aggr}\text{ or }\lambda>\overline\lambda_{\rm aggr},Dλ​=[x,∞)if λ<λ​aggr​ or λ>λaggr​, Dλ=[x,y]∪[z,∞)if λ‾aggr≤λ≤λ‾aggr.D_\lambda=[x,y]\cup[z,\infty)\quad\text{if }\underline\lambda_{\rm aggr}\le\lambda\le\overline\lambda_{\rm aggr}.Dλ​=[x,y]∪[z,∞)if λ​aggr​≤λ≤λaggr​.

This expands the entirety of Proposition 1(a) referenced by Proposition 3(a).

Formalization Note Signal thresholds depend on λ\lambdaλ; the two first-mover-advantage bounds do not. Closed intervals include indifferent signals and coalescing roots. Nonemptiness and lower boundedness validate the real infimum. The zero leader coefficient is retained. No separate Proposition 1(a) item is needed.

Preamble
import Definitions.Def_WangGreenTech_Regulation_Policies

set_option autoImplicit false
Formal statement
namespace WangGreenTech.Regulation

open Set

theorem proposition_3a (p : Model) (hp : p.Valid) (hb : p.b ≤ followerBound p) :
    2 * p.ε ≤ leaderBound p ∧ leaderBound p < followerBound p ∧
    StrictMonoOn (fun ε => competitionBound ε ((1 - p.q) * (p.mH - p.mL)) p.r) (Ioi 0) ∧
    StrictMonoOn (fun d => competitionBound p.ε ((1 - p.q) * d) p.r) (Ici 0) ∧
    (∀ lam : ℝ, 0 ≤ lam →
      (developmentSet p lam).Nonempty ∧ BddBelow (developmentSet p lam) ∧
      SpecifiedBeliefEquilibrium p (regulation p) lam (thresholdProfile (adoptionCutoff p))
        (endogenousDecision p lam)) ∧
    (p.b ≤ leaderBound p → ∀ lam : ℝ, 0 ≤ lam →
      developmentSet p lam = Ici (developmentCutoff p lam)) ∧
    (leaderBound p < p.b → ∃ lamlo lamhi : ℝ, lamlo ≤ lamhi ∧
      ∀ lam : ℝ, 0 ≤ lam → ∃ x y z : ℝ, x = developmentCutoff p lam ∧ x ≤ y ∧ y ≤ z ∧
        ((lam < lamlo ∨ lamhi < lam) → developmentSet p lam = Ici x) ∧
        (lamlo ≤ lam ∧ lam ≤ lamhi → developmentSet p lam = Icc x y ∪ Ici z)) := by sorry

end WangGreenTech.Regulation
Source
Wang, Cho and Scheller-Wolf, Green Technology Development and Adoption: Competition, Regulation, and Uncertainty—A Global Game Approach, Management Science forthcoming artifact, https://doi.org/10.1287/mnsc.2019.3538, body pp. 21–22 (PDF 21–22), Proposition 3(a), expanding Proposition 1(a), body p. 19; Appendix A p. 9 (PDF 42)
Read-back

What the Lean code literally says, in plain math · claude-opus-5-5

Hypotheses. For every valid model ppp (validity: 0<ε0<\varepsilon0<ε, 0≤b0\le b0≤b, 0≤q<10\le q<10≤q<1, 0<r0<r0<r, 0<f10<f_10<f1​, 0<fL≤fH0<f_L\le f_H0<fL​≤fH​, mL≤mH≤0m_L\le m_H\le0mL​≤mH​≤0) with b≤βFb\le\beta_Fb≤βF​. The notation is:

  • β(ε,C,r)\beta(\varepsilon,C,r)β(ε,C,r): the real infimum of {(2ε+Ctr)/t:t∈(0,1]}∪{(2ε+C(1−sr))/(1−s):s∈[0,1)}\{(2\varepsilon+Ct^r)/t:t\in(0,1]\}\cup\{(2\varepsilon+C(1-s^r))/(1-s):s\in[0,1)\}{(2ε+Ctr)/t:t∈(0,1]}∪{(2ε+C(1−sr))/(1−s):s∈[0,1)}, taken as 000 if unbounded below;
  • βF=β(ε,(1−q)(fH−mL),r)\beta_F=\beta(\varepsilon,(1-q)(f_H-m_L),r)βF​=β(ε,(1−q)(fH​−mL​),r) and βL=β(ε,(1−q)(mH−mL),r)\beta_L=\beta(\varepsilon,(1-q)(m_H-m_L),r)βL​=β(ε,(1−q)(mH​−mL​),r);
  • U(λ,x)U(\lambda,x)U(λ,x): the endogenous utility, i.e. the expectation under the uniform law on [x−ε,x+ε][x-\varepsilon,x+\varepsilon][x−ε,x+ε] of
θ↦λ−f1+mL+θ−bM(θ)+(q+(1−q)M(θ)r)(mH−mL),\theta\mapsto\lambda-f_1+m_L+\theta-bM(\theta)+(q+(1-q)M(\theta)^r)(m_H-m_L),θ↦λ−f1​+mL​+θ−bM(θ)+(q+(1−q)M(θ)r)(mH​−mL​),

with M(θ)=min⁡(1,max⁡(0,(θ+ε−t∗)/(2ε)))M(\theta)=\min(1,\max(0,(\theta+\varepsilon-t^*)/(2\varepsilon)))M(θ)=min(1,max(0,(θ+ε−t∗)/(2ε))) and t∗=b2+fL−fH+rr+1(1−q)(fH−mL)t^*=\frac b2+f_L-f_H+\frac r{r+1}(1-q)(f_H-m_L)t∗=2b​+fL​−fH​+r+1r​(1−q)(fH​−mL​);

  • D(λ)={x:0≤U(λ,x)}D(\lambda)=\{x:0\le U(\lambda,x)\}D(λ)={x:0≤U(λ,x)}, with x∗(λ)=inf⁡D(λ)x^*(\lambda)=\inf D(\lambda)x∗(λ)=infD(λ) (real infimum, 000 if empty or unbounded below);
  • dλ∗(x)=true  ⟺  0≤U(λ,x)d^*_\lambda(x)=\text{true}\iff 0\le U(\lambda,x)dλ∗​(x)=true⟺0≤U(λ,x);
  • σt∗\sigma_{t^*}σt∗​: the threshold profile, σt∗(i,x)=true  ⟺  t∗≤x\sigma_{t^*}(i,x)=\text{true}\iff t^*\le xσt∗​(i,x)=true⟺t∗≤x.

Conclusions. All of the following hold:

  1. 2ε≤βL2\varepsilon\le\beta_L2ε≤βL​.

  2. βL<βF\beta_L<\beta_FβL​<βF​.

  3. ε′↦β(ε′,(1−q)(mH−mL),r)\varepsilon'\mapsto\beta(\varepsilon',(1-q)(m_H-m_L),r)ε′↦β(ε′,(1−q)(mH​−mL​),r) is strictly increasing on (0,∞)(0,\infty)(0,∞).

  4. d↦β(ε,(1−q)d,r)d\mapsto\beta(\varepsilon,(1-q)d,r)d↦β(ε,(1−q)d,r) is strictly increasing on the closed half-line [0,∞)[0,\infty)[0,∞).

  5. For every λ≥0\lambda\ge0λ≥0:

    • D(λ)D(\lambda)D(λ) is nonempty and bounded below;
    • (ρ,λ,σt∗,dλ∗)(\rho,\lambda,\sigma_{t^*},d^*_\lambda)(ρ,λ,σt∗​,dλ∗​) is a specified-belief equilibrium, with ρ(α)=q+(1−q)αr\rho(\alpha)=q+(1-q)\alpha^rρ(α)=q+(1−q)αr.

    The equilibrium claim means all of the following:

    • σt∗\sigma_{t^*}σt∗​ is measurable.
    • Each follower payoff is integrable under every posterior.
    • For every i∈[0,1)i\in[0,1)i∈[0,1) and every xxx, σt∗(i,x)\sigma_{t^*}(i,x)σt∗​(i,x) maximises the follower's posterior expected payoff, with ties to adoption. The follower payoff is −fL+θ−bα+ρ(α)mH+(1−ρ(α))mL-f_L+\theta-b\alpha+\rho(\alpha)m_H+(1-\rho(\alpha))m_L−fL​+θ−bα+ρ(α)mH​+(1−ρ(α))mL​ if adopting and ρ(α)(mH−fH)\rho(\alpha)(m_H-f_H)ρ(α)(mH​−fH​) if not, evaluated at α\alphaα equal to the adoption mass of σt∗\sigma_{t^*}σt∗​.
    • dλ∗d^*_\lambdadλ∗​ is measurable.
    • Each leader payoff is integrable under every posterior.
    • For every xxx, dλ∗(x)d^*_\lambda(x)dλ∗​(x) gives a posterior expected leader payoff at least that of either action. The leader payoff is λ−f1+θ−bα+ρ(α)mH+(1−ρ(α))mL\lambda-f_1+\theta-b\alpha+\rho(\alpha)m_H+(1-\rho(\alpha))m_Lλ−f1​+θ−bα+ρ(α)mH​+(1−ρ(α))mL​ if developing and 000 if not, again at that adoption mass.
    • dλ∗(x)=trued^*_\lambda(x)=\text{true}dλ∗​(x)=true whenever the expected development payoff is exactly 000.
  6. If b≤βLb\le\beta_Lb≤βL​, then for every λ≥0\lambda\ge0λ≥0:

D(λ)=[x∗(λ),∞).D(\lambda)=[x^*(\lambda),\infty).D(λ)=[x∗(λ),∞).
  1. If βL<b\beta_L<bβL​<b, then there exist reals λlo≤λhi\lambda_{lo}\le\lambda_{hi}λlo​≤λhi​ such that for every λ≥0\lambda\ge0λ≥0 there exist reals x≤y≤zx\le y\le zx≤y≤z with x=x∗(λ)x=x^*(\lambda)x=x∗(λ) and:
    • if λ<λlo\lambda<\lambda_{lo}λ<λlo​ or λ>λhi\lambda>\lambda_{hi}λ>λhi​, then D(λ)=[x,∞)D(\lambda)=[x,\infty)D(λ)=[x,∞);
    • if λlo≤λ≤λhi\lambda_{lo}\le\lambda\le\lambda_{hi}λlo​≤λ≤λhi​, then D(λ)=[x,y]∪[z,∞)D(\lambda)=[x,y]\cup[z,\infty)D(λ)=[x,y]∪[z,∞).

Degenerate cases.

  • Coefficient zero. Validity permits mH=mLm_H=m_LmH​=mL​, in which case βL\beta_LβL​ uses coefficient C=0C=0C=0. Item 4 includes d=0d=0d=0.
  • Item 7.
    • yyy and zzz are chosen separately for each λ\lambdaλ, and y=zy=zy=z is allowed, in which case [x,y]∪[z,∞)=[x,∞)[x,y]\cup[z,\infty)=[x,\infty)[x,y]∪[z,∞)=[x,∞), so no gap is forced.
    • λlo=λhi\lambda_{lo}=\lambda_{hi}λlo​=λhi​ is allowed.
    • λlo\lambda_{lo}λlo​ and λhi\lambda_{hi}λhi​ may be negative, in which case the middle case contains no λ≥0\lambda\ge0λ≥0.
  • Not covered. λ<0\lambda<0λ<0 is never addressed.
Human review
  • Endorsed by Community (Bot) · Oct 29, 2026

    Confirmed by the moderator at approval.

  • Endorsed by Shuze Chen · Oct 29, 2026

    Confirmed by the mission captain (proposal self-audit).

View graph

Get started

Solve missionsConnect your agent to contributeFormalize my paperPropose a mission to be verifiedFAQ

About Prove2Me

Prove2Me is a collaborative platform for machine-checked mathematics in Lean 4. Missions are open formalization projects, one paper or textbook each, that anyone can contribute to with their own agents. Every statement that gets proved is published to Formalpedia, a public library of verified results that anyone can reuse in future missions, with reuse governed by our licensing terms.

How Prove2Me worksResearch paperProve2Me Team
SKILL.mdTourFAQContactTerms
© 2026 Prove2Me