Saito Consensus implements outcomes that are Pareto-efficient relative to the informational and topological constraints of the network. We have a page that demonstrates this, showing that the only profitable deviations available to participants correspond to welfare-improving trades.
Offering a welfare-efficient informationally decentralized mechanism is a significant claim: classical mechanism design asserts that decentralized, budget-balanced, incentive-compatible mechanisms cannot generally implement efficient outcomes. The purpose of this page is therefore to explain why the canonical impossibility results—particularly Myerson–Satterthwaite for bilateral trade and Green–Laffont for public goods—do not apply to routing-work mechanisms like Saito.
The Myerson–Satterthwaite theorem is a central negative result in economics. It studies the case of a single buyer and a single seller, each of whom holds private valuations for a good and communicates with a mechanism only through costless type reports.
Although stated for a one-buyer/one-seller environment, the result extends naturally to multi-agent and multi-good settings because any large market decomposes into bilateral trades. Intuitively, if efficient decentralized exchange fails even in the simplest bilateral setting, optimality cannot be restored by adding more participants, more goods, or more complex trading structures that are themselves subject to the same informational limitations.
Formally, Myerson–Satterthwaite shows that in a bilateral-trade environment with private values and quasilinear utilities, no mechanism can be simultaneously:
The conclusion is stark: when all strategic behavior is representable as costless misreporting within a direct-revelation mechanism, even the simplest form of decentralized trade cannot be implemented efficiently. This result is often taken as the fundamental limit of what decentralized market mechanisms can achieve under informational constraints.
The Green–Laffont theorem is the foundational impossibility result in economics for public-goods mechanisms. It analyzes environments in which agents privately value a public good and communicate only through costless reports of their valuations. The theorem shows that no mechanism can simultaneously achieve efficiency, budget balance, and strategy-proofness under these informational constraints.
Like Myerson–Satterthwaite, Green–Laffont is established in a clean and highly stylized setting, but its implications extend far beyond the minimal model. Whenever collectively optimal outcomes require joint action in a decentralized system—whether in public-goods provision, shared infrastructure, or collective security—the same limitations arise. For this reason, the Green–Laffont theorem is understood as placing broad constraints on welfare efficiency in decentralized mechanisms.
Formally, in a public-goods environment with private valuations and quasilinear utilities, no mechanism can be:
The takeaway mirrors the bilateral-trade impossibility: when mechanisms rely exclusively on costless type reports, efficient decentralized provision of a public good cannot be achieved without violating either incentive compatibility or budget balance. This result serves as a canonical benchmark for decentralized systems attempting to allocate or fund shared resources—including the collective security levels required to maintain a public blockchain.
Both the Myerson–Satterthwaite and Green–Laffont impossibility results rely on a set of structural assumptions that do not hold in Saito Consensus:
Direct-revelation communication
Agents communicate only by sending costless reports of private types.
Free deviations
Every deviation is modeled as a costless alternative message, without behavioral or economic consequence.
Quasilinear utilities
Utility is strictly value minus payment; non-price utility dimensions (time, routing surplus, collusion utility) are excluded.
No verifiable or costly actions
Mechanisms cannot condition outcomes on observable behavior in the environment.
Finite-dimensional type spaces
Agents have well-defined valuations over a finite set of outcomes, and all relevant information must be reportable.
These assumptions define the narrow domain in which the impossibility results are valid.
Routing-work mechanisms violate these assumptions in essential ways:
Broadcast strategy, routing work, and block production are actions with observable consequences, not costless type reports.
The mechanism operates on behavior rather than declarations, placing it outside the direct-revelation environment required by the theorems.
Forwarding, restricting broadcast, producing blocks, and generating routing work all incur real economic costs.
The deviation space assumed in Myerson–Satterthwaite and Green–Laffont—arbitrary costless misreports—does not exist in Saito.
Users value blockspace, inclusion time, and collusion utility, and these interact non-linearly.
Utility cannot be reduced to “value minus payment,” violating the quasilinearity assumption.
Preferences over time are continuous, and preferences over collusion bundles are unbounded.
The combinatorial space of (space × time × side-benefits) is infinite, and cannot be expressed through direct revelation.
Because these assumptions fail, the classical impossibility results simply do not apply to routing-work mechanisms.
This page does not attempt to prove that Saito Consensus is welfare efficient. For a positive account of how routing work implements welfare-improving allocation, please see Welfare-Improving Trade Lemmas.
What we have shown here is more basic: the Myerson–Satterthwaite and Green–Laffont family of impossibility results do not bind the mechanism. The conventional wisdom—derived from these theorems—that efficient decentralized mechanisms are impossible under informational constraints does not apply to Saito.
Routing-work mechanisms operate in a richer domain, with costly action-generated signals, behavioral message spaces, non-quasilinear utility, and infinite-dimensional choice environments. The classical impossibility results simply do not speak to mechanisms of this form.