Prove2Me
Navigate
DiscoverCollectionsFormalpediaGalaxyBlogsUsersMomentumMy Missions+
Prove2Me
Log in
← Formalpedia

Proposition 2, p. 12 — the optimal transaction benefit b* decreases in k and is constant, then decreasing, then increasing in μ; leakage is zero, then increasing, then decreasing

Proved
MarketplaceLeakage.Benefits.proposition_2

by mikedeng1 · Oct 15, 2026 · Mathlib 0df444a (Lean v4.33.1)

comparative-staticsinvestmentleakagep2o-batch-pfp2ap2o-gran-per-chapterp2o-plan-paperp2o-v1platforms

Consider the leakage game with transaction benefits (see the Setting): switching costs are uniform on [0,μ][0,\mu][0,μ], μ>0\mu > 0μ>0, the marketplace invests K(b)=k2b2K(b) = \tfrac{k}{2}b^2K(b)=2k​b2 with k>0k > 0k>0 in a benefit b≥0b \ge 0b≥0 and charges a fee b≤f≤v−c+bb \le f \le v - c + bb≤f≤v−c+b, and the seller (c<vc < vc<v) answers with a profit-maximizing direct price pd≤vp_d \le vpd​≤v. Let

μˉ=v−c2+14k+(v−c2)2+116k2.\bar\mu = \frac{v-c}{2} + \frac{1}{4k} + \sqrt{\Bigl(\frac{v-c}{2}\Bigr)^2 + \frac{1}{16k^2}} .μˉ​=2v−c​+4k1​+(2v−c​)2+16k21​​.

Then:

  1. An optimum (b∗,f∗,pd∗)(b^*, f^*, p_d^*)(b∗,f∗,pd∗​) exists, and at every optimum the marketplace's optimal choice of transaction benefits is
b∗={1kif μ≤12k,12k−12μif 12k≤μ≤μˉ,1k−v−c2kμif μ≥μˉ.b^* = \begin{cases} \dfrac{1}{k} & \text{if } \mu \le \tfrac{1}{2k}, \\[6pt] \dfrac{1}{2k - \frac{1}{2\mu}} & \text{if } \tfrac{1}{2k} \le \mu \le \bar\mu, \\[6pt] \dfrac{1}{k} - \dfrac{v-c}{2k\mu} & \text{if } \mu \ge \bar\mu. \end{cases}b∗=⎩⎨⎧​k1​2k−2μ1​1​k1​−2kμv−c​​if μ≤2k1​,if 2k1​≤μ≤μˉ​,if μ≥μˉ​.​
  1. b∗b^*b∗ is strictly decreasing in the investment cost kkk: for fixed μ\muμ and k1<k2k_1 < k_2k1​<k2​, every optimal investment at k2k_2k2​ is strictly smaller than every optimal investment at k1k_1k1​.
  2. For fixed kkk, as a function of the switching-cost scale μ\muμ: for μ≤1/(2k)\mu \le 1/(2k)μ≤1/(2k) the optimal investment is constant at 1/k1/k1/k and the equilibrium leakage G(v−pd∗)G(v - p_d^*)G(v−pd∗​) is zero; on [1/(2k),μˉ][1/(2k), \bar\mu][1/(2k),μˉ​] the optimal investment is strictly decreasing and the leakage strictly increasing in μ\muμ; on [μˉ,∞)[\bar\mu, \infty)[μˉ​,∞) the optimal investment is strictly increasing and the leakage strictly decreasing in μ\muμ.

The proposition shows that the marketplace's investment in fighting leakage is non-monotone in switching costs: with low switching costs it buys out leakage entirely; with intermediate ones it tolerates growing leakage and invests less; with high ones it again invests more as leakage falls.

Formalization Note "Optimal" means an optimum of the joint problem over (b,f)(b,f)(b,f) with the seller best-responding (IsOpt); no closed form for b∗b^*b∗ is defined. The comparative statics compare optimal choices at two parameter values. "Decreasing" and "increasing" are read strictly, as on the page. The cases of b∗b^*b∗ agree at μ=1/(2k)\mu = 1/(2k)μ=1/(2k) and μ=μˉ\mu = \bar\muμ=μˉ​, so the boundary convention of the Lean if is immaterial.

Preamble
import Mathlib
import Definitions.Def_MarketplaceLeakage_Benefits_Setting
Formal statement
namespace MarketplaceLeakage.Benefits

/-- Proposition 2, p. 12: with uniform switching costs on `[0, μ]` and `K(b) = (k/2) b²`, `k > 0`,
the marketplace's optimal transaction benefit is `b* = 1/k` for `μ ≤ 1/(2k)`,
`1/(2k − 1/(2μ))` for `1/(2k) ≤ μ ≤ μ̄` and `1/k − (v − c)/(2kμ)` for `μ ≥ μ̄`; `b*` is
strictly decreasing in `k`; in `μ` it is constant, then strictly decreasing on `[1/(2k), μ̄]`,
then strictly increasing on `[μ̄, ∞)`; the equilibrium leakage is zero, then strictly increasing
on `[1/(2k), μ̄]`, then strictly decreasing on `[μ̄, ∞)`. -/
theorem proposition_2 (v c : ℝ) (hvc : c < v) :
    -- the optimal investment b*: an optimum exists, and every optimum has b = b*
    (∀ μ k : ℝ, 0 < μ → 0 < k →
      (∃ b f pd : ℝ, IsOpt v c μ k b f pd) ∧
      ∀ b f pd : ℝ, IsOpt v c μ k b f pd →
        b = (if μ ≤ 1 / (2 * k) then 1 / k
             else if μ ≤ muBar v c k then 1 / (2 * k - 1 / (2 * μ))
             else 1 / k - (v - c) / (2 * k * μ))) ∧
    -- b* is strictly decreasing in the investment cost k
    (∀ μ k₁ k₂ : ℝ, 0 < μ → 0 < k₁ → k₁ < k₂ → ∀ b₁ f₁ p₁ b₂ f₂ p₂ : ℝ,
      IsOpt v c μ k₁ b₁ f₁ p₁ → IsOpt v c μ k₂ b₂ f₂ p₂ → b₂ < b₁) ∧
    -- dependence of b* and of the MarketplaceLeakage.Baseline.leakage on μ, for fixed k
    (∀ k : ℝ, 0 < k →
      (∀ μ : ℝ, 0 < μ → μ ≤ 1 / (2 * k) → ∀ b f pd : ℝ, IsOpt v c μ k b f pd →
        b = 1 / k ∧ MarketplaceLeakage.Baseline.leakage v μ pd = 0) ∧
      (∀ μ₁ μ₂ : ℝ, 1 / (2 * k) ≤ μ₁ → μ₁ < μ₂ → μ₂ ≤ muBar v c k →
        ∀ b₁ f₁ p₁ b₂ f₂ p₂ : ℝ, IsOpt v c μ₁ k b₁ f₁ p₁ → IsOpt v c μ₂ k b₂ f₂ p₂ →
          b₂ < b₁ ∧ MarketplaceLeakage.Baseline.leakage v μ₁ p₁ < MarketplaceLeakage.Baseline.leakage v μ₂ p₂) ∧
      (∀ μ₁ μ₂ : ℝ, muBar v c k ≤ μ₁ → μ₁ < μ₂ →
        ∀ b₁ f₁ p₁ b₂ f₂ p₂ : ℝ, IsOpt v c μ₁ k b₁ f₁ p₁ → IsOpt v c μ₂ k b₂ f₂ p₂ →
          b₁ < b₂ ∧ MarketplaceLeakage.Baseline.leakage v μ₂ p₂ < MarketplaceLeakage.Baseline.leakage v μ₁ p₁)) := by sorry

end MarketplaceLeakage.Benefits
Source
Hagiu, Wright, Marketplace Leakage, authors' manuscript dated March 2023 (sha256 cb89c…98a6; published in Management Science, doi:10.1287/mnsc.2023.4757), p. 12, Proposition 2
Human review
  • Endorsed by Shuze Chen · Oct 15, 2026

    Confirmed by the moderator at approval.

  • Endorsed by mikedeng1 · Oct 15, 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