The Third Court: Insurance, Arbitration, and the Private Production of Justice

A merchant court can settle your contract dispute by afternoon, and a victim-centered legal tradition can price the injury your assailant owes you in camels or coin. I have made both arguments in previous posts. Neither argument, taken alone, answers the critic who replies: and then what? The merchant court cannot prevent the crime that precedes the dispute. The restitution system cannot enforce its own rulings against a defiant offender who moves to the next town. Each institution solves one part of the problem while quietly assuming that something else will solve the rest. That something else has always been the state, which is why every partial critique of the state's legal monopoly eventually hits the same wall. You can disassemble the monopoly piece by piece, and your critic will simply point to the piece you have not yet reached.

The answer is to stop presenting the pieces separately and start describing the machine they compose. Insurance, arbitration, restitution, and cryptographic enforcement are not four independent proposals for replacing fragments of state justice. They are four layers of a single system, and the system works because each layer feeds the others in ways that no layer achieves alone.

Begin with the layer that historical private law always lacked and that modern proposals have yet to take seriously enough: insurance. The piepowder court resolved disputes brilliantly, but it did nothing to prevent them. The Icelandic Althing priced injuries after they occurred, but it could not change the calculus of a man deciding whether to commit one. Insurance operates before the injury and governs behavior through price.

An insurer covering a community against theft and violence profits when claims do not occur. Dwell on that sentence, because its implications run deeper than they first appear. Every state institution nominally tasked with providing security operates under the opposite incentive. Police departments whose jurisdictions grow more dangerous receive larger budgets, which means rising crime is good for the department even as it is catastrophic for the people the department claims to protect. Prosecutors build careers on conviction counts, and a world without crime would be a world without prosecutors. The prison industry expands its revenue by housing more inmates, and since seventy percent of those inmates will return within a few years of release, the industry's best customers are its own failures. An insurance company embedded in this same population would calculate every number in reverse. Every assault prevented is a claim that never arrives. Every conflict mediated before it escalates is a lawsuit that never generates payout. The insurer who reduces crime in his coverage area does not lose his budget. He widens his margin.

The mechanism is premium pricing, and premium pricing accomplishes what criminal codes attempt with a precision that legislation cannot match. A statute says that theft carries up to five years imprisonment, and the penalty is the same whether the thief targeted a hardened vault or an unlocked car, whether the neighborhood is dangerous or quiet, whether the offender has committed a dozen prior offenses or none. Legislation is a blunt instrument designed for categories. An insurer prices every variable that affects risk: the quality of your locks, the lighting on your street, the claims history of your neighborhood, the company you keep. A client surrounded by low-risk neighbors in a well-secured area pays less. A client whose associations and behavior elevate the insurer's exposure pays more, and if the risk exceeds what any premium can cover, the insurer declines coverage entirely. No one is imprisoned. No armed enforcer appears. The behavioral governance that criminal law promises and fails to deliver arrives through the quarterly premium notice, calibrated to the individual case with an actuarial precision that no judge possesses and no legislature could encode.

This pricing mechanism also produces something that tax-funded policing structurally cannot: a real price signal for security. When a city allocates police patrol routes, the data comes from the same department seeking a budget increase, filtered through political calculations that have nothing to do with where protection is most needed. An insurance company allocates defensive resources according to where claims actually arise, because every misallocation comes out of the insurer's own revenue. Actuarial tables do not lobby. They report what happened, and the insurer who disregards them goes bankrupt. The same information problem that prevents central planners from allocating steel or wheat efficiently prevents centralized police bureaucracies from allocating patrol hours efficiently, and it fails for the same reason: the relevant knowledge is dispersed across millions of individual circumstances that no planning bureau can collect, but that competitive pricing aggregates automatically.

Insurance also dissolves the coordination problem that critics raise whenever private defense is proposed. How do atomized individuals organize collectively against threats without recreating the monopoly they meant to escape? They do not need to organize at all; they need to buy coverage. The insurer organizes on their behalf because disorganization is expensive: a neighborhood without patrol services and mediation resources generates claims that destroy profitability. Each household purchases protection for private reasons, and the insurer coordinates collective defense as an unplanned consequence of managing its own risk exposure. The frankpledge tithing in Anglo-Saxon England worked on the same logic a millennium ago, binding ten households into mutual liability for each member's conduct so that every man had a financial stake in his neighbor's behavior. Friendly societies in Britain scaled the principle to six million members by 1910. Lloyd's of London began as a coffee house and over three centuries built governance structures and claim settlement procedures that regulated maritime safety worldwide without a single statute requiring participation. These institutions did not theorize about insurance as governance. They practiced it, and the practice worked because the incentive to minimize losses disciplines behavior more reliably than any criminal code.

When prevention fails and a dispute arises, the second layer activates: arbitration. The insurer has already specified in the coverage contract which arbitration firms will hear disputes and under what procedural rules. Different insurers offer different frameworks. Some emphasize speed, others thoroughness. Some require mediation before adjudication, others move directly to binding resolution. The client who needs disputes settled by close of business chooses an insurer whose arbitration agreements reflect that preference. The client who demands exhaustive procedural protections chooses differently, and no legislature deliberates over the optimal procedure. Procedural diversity arises from the same competitive process that produces diversity in every other market: suppliers differentiate and customers sort themselves according to their own valuations of what a legal process should look like.

Arbitration generates its own body of law through the evolutionary process that the medieval Law Merchant proved over seven centuries. Arbitrators who resolve disputes repeatedly within a trade community accumulate precedent. Customs that facilitate cooperation survive and spread. Rules that obstruct commerce die out as parties abandon the arbitrators who apply them. That same evolutionary process operates today in commercial arbitration, where the American Arbitration Association processes over 200,000 cases annually and ninety percent of international commercial contracts already include arbitration clauses. Commerce has voted with its wallets by overwhelmingly choosing private resolution over state courts. The monopolist retains exactly the customer profile that monopolists always retain: those too poor or too uninformed to leave.

When clients of different insurers dispute, their insurers invoke pre-negotiated inter-insurer arbitration agreements, the same mechanism by which Visa and Mastercard developed chargeback procedures and by which every network industry in history has resolved cross-network conflicts. The arbitrator selected through these agreements earns revenue from both insurers, which means her reputation for fairness and competence is her capital. You cannot fire the judge assigned to your case in a state court, and you cannot take your business to a competing jurisdiction without physically relocating. A private arbitrator who acquires a reputation for bias or slowness loses cases to competitors and eventually loses the inter-insurer agreements that supply her caseload. Accountability follows from the same competitive pressure that disciplines every other service provider in a market economy.

The third layer is restitution, and it redirects the remedy to the only party who deserves it: the victim. When an arbitrator rules, payment flows to the person who was harmed, not to a treasury that will spend it on the salaries of the bureaucrats who processed the case. The victim owns the claim from the moment of injury, and ownership means the claim carries every property of any other economic asset. It can be valued. It can be sold. A victim who lacks the resources to pursue a powerful offender can transfer the claim to a collection firm or a champion willing to invest in prosecution for a share of the recovery, precisely as a farmer wronged by a chieftain in medieval Iceland could sell his right to collect damages to another chieftain who had both the resources and the motive to pursue it. That practice sustained a legal system for over three centuries and solved the problem that state criminal justice claims to address through public prosecution but actually addresses by granting prosecutors the discretion to ignore victims whose cases fail to advance institutional priorities.

Restitution realigns every incentive that imprisonment destroys. An offender who owes a debt to the person he injured has a reason to work: repayment is the path back to full standing in the community. An offender warehoused in a cell has no such reason. Imprisonment costs tens of thousands of dollars per year and produces nothing for anyone. Victims receive no compensation while offenders learn criminal technique from cellmates and emerge less employable and more dangerous than when they entered, so that the community bears first the cost of confinement and then the cost of the recidivism that confinement produces. Under restitution, the offender's labor generates value directed to the person actually harmed while the community avoids the entire expense of caging him. Rothbard put the principle with characteristic directness: the criminal loses his rights to the extent that he has invaded the rights of another, and the emphasis in punishment must fall on paying the debt to the victim. Human beings respond to incentives. A system that makes productive repayment the road to restored standing produces different behavior than a system that makes caged idleness the mandatory response to every offense.

The fourth layer completes the circuit that every historical private law system left open. Medieval merchant courts enforced their rulings through reputation and exclusion: a trader who defied a piepowder court found that no merchant at the fair would deal with him, and the Hanseatic League maintained blacklists so effective that any port harboring a blacklisted merchant risked its own standing with the League. Beth Din enforced judgments through community standing that could follow a defiant member across generations. These mechanisms worked within bounded communities where everyone knew everyone. They could not scale to anonymous global commerce where counterparties have never met and may never interact again.

Cryptographic enforcement removes this limitation. A Nostr web of trust produces a public, auditable reputation record that any counterparty anywhere on earth can inspect before entering a deal. The trader whose record shows consistent compliance with arbitration rulings and prompt payment of restitution obligations commands better terms and lower insurance premiums. The trader whose record shows defiance and unpaid debts finds that counterparties demand higher escrow deposits while insurers charge premiums reflecting the elevated risk. Bitcoin multisig escrow makes the enforcement mechanical: two of three keys must sign to release funds, and the arbitrator who holds the deciding key participates only when the parties cannot agree. Discreet Log Contracts lock funds on-chain and release them according to oracle outcomes, so that an arbitrator's ruling triggers payment automatically, with no enforcement agent required and no possibility of the losing party absconding with funds already committed.

What matters most is not the individual layers but how they feed each other. Insurance funds arbitration by specifying resolution procedures in coverage contracts and paying arbitrators from premium revenue. Arbitration orders restitution according to standards that evolve through competitive selection. Restitution payments flow through cryptographic channels that the losing party committed to at the time of the original transaction. Enforcement feeds back into insurance: a client who defies an arbitration ruling sees his premiums rise to reflect the increased risk he represents, and if defiance persists, the insurer terminates coverage. In a system where coverage is the passport to commerce, termination is the modern equivalent of expulsion from a medieval fair. The circle closes, and each layer strengthens every other.

The composition also creates a redundancy that the state monopoly cannot match. If an insurer fails, clients switch to competitors. Should an arbitrator prove corrupt, the inter-insurer agreements that supplied his caseload migrate to alternatives. Even where a particular enforcement mechanism is circumvented, the remaining layers still function: a trader who evades Bitcoin escrow still faces reputation damage that affects future insurance pricing and future access to trading partners. No single point of failure can bring the system down because no single institution controls any layer. The state monopoly is fragile in exactly the ways this system is not: a corrupt judge cannot be fired by the litigants, a police department that clears fewer than half of violent crimes cannot lose its customers to a competitor, and a prison system producing seventy-percent recidivism continues to receive tax funding because the taxpayers have no alternative provider and no way to exit.

Critics will raise the warlord scenario. Without a monopoly enforcer, the strongest private agency imposes its will and becomes a de facto state. The objection is a description of what states already are. Every existing government began as a warlord that conquered territory and declared a monopoly on force. The question is whether competitive institutions would produce fewer and weaker warlords than the arrangement that gave the twentieth century its body count. Under competition, an aggressive insurer would watch its claims skyrocket as the violence it initiated destroyed the property it had contracted to protect. Its premiums would become uncompetitive as costs rose, and its clients would flee to peaceful competitors because, unlike citizens of a state, they are not trapped within borders enforced by the very monopolist they wish to escape. Other insurers would refuse to honor inter-insurer agreements with the aggressor, severing it from the arbitration network and the reputation infrastructure that commerce requires. The economic logic that makes cartels unstable in commodity markets operates with equal force in the market for defense.

What emerges from these four layers is not a utopia but a system whose failures are local and correctable. At every point where the state monopoly insulates failure from consequence, the private stack connects failure to loss. Loss is the only signal that drives correction, and it is the one signal the monopolist never receives.

I have not described four speculative proposals. Merchant courts resolved commercial disputes at Champagne fairs eight centuries ago. Victim restitution sustained legal systems from Iceland to Somalia for longer than the American constitutional order has existed. Insurance mutuals governed behavior from Anglo-Saxon tithings to Victorian friendly societies. Cryptographic enforcement is the newest addition, and it is the piece that binds the others into a global system no longer dependent on bounded communities where everyone knows everyone. Each layer was proven separately across centuries and civilizations. What remains is the recognition that they compose into something greater than any of them achieved alone: a complete private legal system where insurance deters, arbitration resolves, restitution heals, and reputation ensures that compliance follows from calculated self-interest across any distance between parties who have never met and may never meet again.