We present a strategy-proof public goods budgeting mechanism where agents determine both the total volume of expanses and the specific allocation. It is constructed as a modification of VCG to a less typical environment, namely where we do not assume quasi-linear utilities nor direct revelation. We further show that under plausible assumptions it satisfies strategy-proofness in strictly dominant strategies, and consequently implements the social optimum as a Coalition Proof Nash Equilibrium. A primary (albeit not an exclusive) motivation of our model is Participatory Budgeting, where members of a community collectively decide the spending policy of public tax dollars. While incentives alignment in our mechanism, as in classic VCG, is achieved via individual payments we charge from agents, in a PB context that seems unreasonable. Our second main result thus provides that, under further specifications relevant in that context, these payments will vanish in large populations. In the last section we expand the mechanism's definition to a class of mechanisms in which the designer can prioritize certain outcomes she sees as desirable. In particular we give the example of favoring equitable (egalitarian) allocations.

翻译：我们提出了一种防策略的公共品预算机制，其中参与者既决定总支出规模也决定具体分配方案。该机制通过将VCG机制改造适用于非典型环境——即不假设拟线性效用函数或直接显示偏好——而构建。我们进一步证明，在合理假设下该机制满足严格占优策略防策略性，并因此将社会最优实现为联盟证明纳什均衡。该模型的主要（尽管非唯一）动机来自参与式预算，即社区成员集体决定公共税款支出政策。虽然我们的机制如同经典VCG一样通过向参与者收取个体费用实现激励相容，但在参与式预算背景下这似乎不合理。因此我们的第二个主要结果表明，在该背景相关的进一步规范条件下，这些费用将在大规模群体中消失。在最后一节中，我们将该机制的定义扩展到一类机制，使得设计者能够优先考虑其认为理想的结果。特别地，我们给出了偏好公平（平等主义）分配的示例。