Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 1, p. 10 — in any financial network, bank i's optimal risk exposure is increasing in every other bank's Z_m

Proved
RiskExposure.Complementarity.proposition_1

by mikedeng1 · Oct 14, 2026 · Mathlib 0df444a (Lean v4.33.1)

financial-networksp2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1strategic-complementaritysupermodular-game

Consider NNN banks with interbank liabilities dˉij≥0\bar d_{ij}\ge0dˉij​≥0 (bank jjj owes bank iii), all totals dˉj>0\bar d_j>0dˉj​>0, deposits vi>0v_i>0vi​>0, a project set [Z‾,Z‾][\underline Z,\overline Z][Z​,Z] with Z‾≥vi+dˉi\underline Z\ge v_i+\bar d_iZ​≥vi​+dˉi​ for every iii (no systemic default), and a success probability PPP with P(z)∈(0,1)P(z)\in(0,1)P(z)∈(0,1), PPP decreasing and P(z) zP(z)\,zP(z)z concave on [Z‾,Z‾][\underline Z,\overline Z][Z​,Z] (Assumption 1). Payments clear at the greatest payment equilibrium, and Z^i(Z−i)=argmax⁡Zi∈[Z‾,Z‾]E[Πi(ω;Z)]\hat Z_i(Z_{-i})=\operatorname{argmax}_{Z_i\in[\underline Z,\overline Z]}\mathbb E[\Pi_i(\omega;Z)]Z^i​(Z−i​)=argmaxZi​∈[Z​,Z]​E[Πi​(ω;Z)] is bank iii's best-response set.

Let i≠mi\ne mi=m be banks, Z∈[Z‾,Z‾]NZ\in[\underline Z,\overline Z]^NZ∈[Z​,Z]N, Zm′∈[Z‾,Z‾]Z'_m\in[\underline Z,\overline Z]Zm′​∈[Z​,Z] with Zm≤Zm′Z_m\le Z'_mZm​≤Zm′​, and Z′Z'Z′ equal to ZZZ with ZmZ_mZm​ replaced by Zm′Z'_mZm′​. If z^∈Z^i(Z−i)\hat z\in\hat Z_i(Z_{-i})z^∈Z^i​(Z−i​) and z^′∈Z^i(Z−i′)\hat z'\in\hat Z_i(Z'_{-i})z^′∈Z^i​(Z−i′​), then

max⁡(z^,z^′)∈Z^i(Z−i′)andmin⁡(z^,z^′)∈Z^i(Z−i).\max(\hat z,\hat z')\in\hat Z_i(Z'_{-i})\qquad\text{and}\qquad\min(\hat z,\hat z')\in\hat Z_i(Z_{-i}).max(z^,z^′)∈Z^i​(Z−i′​)andmin(z^,z^′)∈Z^i​(Z−i​).

In words: in any financial network, regular or not, risk exposures are strategic complements — bank iii's optimal risk exposure is nondecreasing in every other bank's risk exposure.

Formalization Note The paper writes dZ^i/dZm≥0d\hat Z_i/dZ_m\ge0dZ^i​/dZm​≥0, which presupposes a unique, differentiable maximizer; Assumption 1 gives neither. The statement above is the strong-set-order form of "Z^i\hat Z_iZ^i​ is nondecreasing in ZmZ_mZm​" (Topkis; Milgrom–Shannon). It implies z^≤z^′\hat z\le\hat z'z^≤z^′ whenever both best-response sets are singletons, and dZ^i/dZm≥0d\hat Z_i/dZ_m\ge0dZ^i​/dZm​≥0 for any differentiable selection. It is vacuous when a best-response set is empty, as the paper's Z^i\hat Z_iZ^i​ presupposes a maximizer. No regularity or path-connectedness of the network is assumed, and self-loops dˉii>0\bar d_{ii}>0dˉii​>0 are allowed.

Preamble
import Mathlib
import Definitions.Def_RiskExposure_Complementarity_Model
Formal statement
namespace RiskExposure.Complementarity

theorem proposition_1 {ι : Type*} [Fintype ι] [DecidableEq ι]
    (dbar : ι → ι → ℝ) (hdbar : ∀ i j, 0 ≤ dbar i j) (hdbarTot : ∀ j, 0 < dbarTot dbar j)
    (v : ι → ℝ) (hv : ∀ i, 0 < v i)
    (Zlo Zhi : ℝ) (hZ : Zlo ≤ Zhi) (hnd : ∀ i, v i + dbarTot dbar i ≤ Zlo)
    (P : ℝ → ℝ) (hP01 : ∀ z ∈ Set.Icc Zlo Zhi, 0 < P z ∧ P z < 1)
    (hPanti : AntitoneOn P (Set.Icc Zlo Zhi))
    (hPconc : ConcaveOn ℝ (Set.Icc Zlo Zhi) (fun z => P z * z))
    (i m : ι) (hmi : m ≠ i) (Z : ι → ℝ) (hZbox : ∀ j, Z j ∈ Set.Icc Zlo Zhi)
    (zm' : ℝ) (hzm' : zm' ∈ Set.Icc Zlo Zhi) (hle : Z m ≤ zm')
    (zh zh' : ℝ) (hzh : zh ∈ bestResponse dbar v P Zlo Zhi Z i)
    (hzh' : zh' ∈ bestResponse dbar v P Zlo Zhi (Function.update Z m zm') i) :
    max zh zh' ∈ bestResponse dbar v P Zlo Zhi (Function.update Z m zm') i ∧
      min zh zh' ∈ bestResponse dbar v P Zlo Zhi Z i := by sorry

end RiskExposure.Complementarity
Source
Shu, Endogenous Risk Exposure and Systemic Instability, Management Science, doi:10.1287/mnsc.2022.01519 (accepted manuscript, 17 Sep 2024), p. 10, Proposition 1; proof Internet Appendix pp. 7–10
Human review
  • Endorsed by Shuze Chen · Oct 14, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 14, 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