Table of Contents
The lawyer slides a document across the mahogany desk. Fifty pages of clauses, subclauses, and contingencies for every imaginable scenario. The business partner nods approvingly. Everyone feels protected. Three months later, something happens that nobody thought to put in the contract. Now the real negotiation begins.
This scene plays out in boardrooms everywhere. We assume more pages equal more security. We believe precise language prevents problems. Game theory suggests otherwise.
The Paradox of Perfect Planning
Traditional contract theory operates on a simple premise. Write down all possible futures. Specify actions for each outcome. Sign on the dotted line. Problem solved.
Except the future refuses to fit neatly into predetermined boxes.
Consider two companies forming a partnership. They could spend months drafting agreements covering supplier failures, market crashes, regulatory changes, and technological disruptions. They could define dispute resolution procedures for scenarios A through Z. The document might be comprehensive, legally sound, and completely useless when scenario AA arrives unannounced.
Game theorists call this the incomplete contracts problem. The name understates the situation. All contracts are incomplete. The question is whether we admit it.
Nobel laureate Oliver Hart built his career studying why parties cannot write complete contracts. The reasons sound obvious once stated. The future contains infinite possibilities. Language fails to capture every nuance. Even identifying all relevant scenarios costs more than the contract is worth. And critically, courts cannot verify many important actions or outcomes.
This last point deserves attention. A contract means nothing if nobody can enforce it. Imagine writing a clause that says “Party A must exert maximum effort.” What does maximum effort look like? How does a judge measure it? The contractor claims they worked hard. The client disagrees. The contract offers no resolution because effort remains invisible to outside observers.
The Game Everyone Plays
Enter game theory with a different framework. Instead of viewing contracts as comprehensive rulebooks, consider them as establishing starting conditions for an ongoing game between parties.
The repeated prisoner’s dilemma illustrates this beautifully. Two parties face ongoing decisions about whether to cooperate or defect. Each round, cooperation benefits both players collectively. But individually, each player gains more by defecting while the other cooperates.
Write this interaction into a contract and problems emerge immediately. How do you define cooperation precisely? What counts as defecting? A million judgment calls resist reduction to legal language.
Yet people cooperate successfully all the time without contracts. The secret lies not in specifying every action but in creating conditions where cooperation becomes the rational choice. The game theory term is “shadow of the future.” When parties expect to interact repeatedly, today’s betrayal costs tomorrow’s cooperation.
This insight flips conventional wisdom. The best agreements might be the shortest ones. Not because details don’t matter, but because rigid specifications can destroy the flexibility that makes partnerships work.
When Lawyers Make Things Worse
A software company needs a consultant for six months. They could write a contract specifying exactly which tasks the consultant performs each week, how many hours constitute reasonable effort, what quality standards apply, and how to handle every conceivable problem.
Or they could write three sentences. The project. The payment. The timeline.
The detailed contract feels safer. It also signals distrust. Every clause that specifies consequences for poor performance tells the consultant: “We assume you’ll try to cheat us.” The consultant reads between the lines and adjusts their mental model of the relationship accordingly.
Game theory research shows that cheap talk isn’t cheap at all. The messages parties send before playing a game significantly affect outcomes. A detailed contract sends a clear message. This is an adversarial relationship. Protect yourself.
That message becomes self-fulfilling. Consultants who feel mistrusted invest less emotional energy in the project’s success. Why go beyond the specified minimum when the client already assumes the worst? The contract designed to prevent sidestepping actually creates it.
Meanwhile, the three sentence agreement signals something else. We believe you’ll do good work. We expect this relationship to benefit both of us. That signal also becomes self-fulfilling. Consultants trusted to exercise judgment often perform better than those operating under surveillance.
This isn’t naive optimism. It’s strategic trust. The parties understand that formal monitoring is expensive and incomplete. They choose instead to invest in relationship capital that encourages cooperation through reputation effects rather than legal enforcement.
The Restaurant Test
Restaurants offer a daily experiment in incomplete contracts. The customer orders a meal. The implicit contract seems simple. Restaurant provides food. Customer pays money. Done.
Except nothing is actually specified. How large is the portion? How long until the food arrives? What happens if the customer dislikes it? What if they find a hair in the soup? The menu lists prices but says nothing about these crucial details.
Some restaurants try to make contracts more complete. They add policies. “Management reserves the right to refuse service.” “Parties of six or more include automatic gratuity.” “No substitutions.” Each rule represents an attempt to specify contingencies.
Other restaurants rely entirely on unwritten norms. They trust servers to use judgment. They empower managers to resolve disputes case by case. They build reputation instead of rules.
Which approach works better? Game theory suggests the answer depends on whether parties expect repeated interaction. Chain restaurants serving highway travelers lean toward explicit policies. Neighborhood spots cultivating regular customers lean toward flexibility and trust.
The repeated game changes everything. Customers who expect to return care about reputation. Restaurants that want loyal patrons care about goodwill. Both sides have incentive to be reasonable even when formal rules don’t specify the outcome.
This pattern appears everywhere once you notice it. Successful marriages run on trust, not prenuptial agreements covering every possible scenario. Lasting friendships operate without contracts defining acceptable behavior. Effective teams coordinate through shared understanding rather than detailed protocols.
The Verification Problem
Here’s where incomplete contracts theory gets interesting. Even when parties can imagine future scenarios, they often cannot write enforceable terms about them.
A publisher signs an author for a book. The contract could specify that the author must “write a good book.” But good according to whom? The publisher’s marketing team? Literary critics? Sales figures? Each measure captures something real yet none provides an objective standard a court could enforce.
So contracts resort to verifiable metrics. Word count. Delivery date. Factual accuracy. These elements can be checked and enforced. But they miss the essential quality that makes a book succeed.
Game theory calls these verifiable versus unverifiable actions. The verifiable stuff gets written into contracts. The unverifiable stuff, which often matters more, depends on trust and relationship dynamics.
This creates an unexpected problem. Detailed contracts focus attention on verifiable metrics at the expense of unverifiable ones. The author hits the word count by adding filler. They meet the deadline by sacrificing quality. The contract specified the wrong things because it could only specify verifiable things.
An incomplete contract acknowledging its own limits sometimes works better. “Write a book both parties consider successful.” This clause is legally weak but strategically strong. It forces ongoing communication. It preserves flexibility. It keeps focus on the actual goal rather than proxy metrics.
When Trust Fails
The case for trust needs honest limitations. Game theory isn’t romantic. It analyzes incentives coldly.
Trust works in repeated games where reputation matters. It works when parties value future interactions. It works in communities where information travels and bad behavior gets punished.
Trust fails in one shot interactions. It fails when parties can disappear after defecting. It fails when information asymmetries let people hide bad behavior.
Buying a used car from a stranger demonstrates this perfectly. No repeated interaction. No shared community. High incentive to misrepresent the vehicle’s condition. This scenario calls for detailed contracts, inspections, and warranties. Trust alone is foolish.
The incomplete contracts framework doesn’t replace formal agreements. It suggests that contracts and trust play complementary roles. Contracts establish baseline expectations and handle verifiable outcomes. Trust manages the unverifiable remainder.
Smart parties understand which situations require which approach. Lending money to a stranger requires contracts. Working with a longtime business partner requires trust. Most real situations fall somewhere between these extremes.
The Renegotiation Dance
Something changes. The partnership that made perfect sense last year now requires adjustment. If the contract specified everything, renegotiation becomes a legal battle over whether circumstances justify modification.
Incomplete contracts build in the expectation of renegotiation. They acknowledge that parties will need to adapt. This sounds like weakness but functions as strength.
Game theory shows that the possibility of renegotiation can actually increase initial cooperation. When parties know they can adjust terms later, they worry less about being locked into bad deals. This reduces defensive behavior at the outset.
A construction contract offers a clear example. Detailed specifications might require using specific materials. But what if better materials become available mid project? A complete contract makes adaptation difficult. An incomplete contract preserving flexibility allows both parties to benefit from improvements.
The key insight is that incomplete contracts shift risk in a particular direction. Complete contracts try to eliminate uncertainty through specification. Incomplete contracts accept uncertainty and create mechanisms to handle it through ongoing cooperation.
Which approach is more efficient? The answer depends on how costly various risks are and how well parties can handle them through relationship management versus legal enforcement.
The Venture Capital Model
Venture capitalists write famously incomplete contracts. They invest millions based on vague business plans. The startup promises to build something valuable. The VC promises to provide capital and advice. Many crucial details remain unspecified.
This looks insane until you understand the game. Both parties know the startup will pivot multiple times. The original plan will prove wrong. Success requires constant adaptation to new information.
A complete contract trying to specify the startup’s obligations would either be so vague as to be meaningless or so specific as to prevent necessary pivots. Instead, VCs rely on control rights, board seats, and reputation mechanisms.
The incomplete contract acknowledges reality. We cannot predict what this company should do in 18 months. We can establish that we’re partners in figuring it out. The contract creates a framework for ongoing cooperation rather than trying to predetermine all outcomes.
Game theory explains why this works better than detailed requirements. The startup has private information about technical possibilities and market conditions. The VC has private information about fundraising environments and strategic options. Both pieces of information evolve constantly.
Trying to specify actions based on contingencies requires sharing all this private information upfront and predicting its evolution. Impossible. Instead, the incomplete contract preserves incentives for parties to share information as it arrives and make decisions jointly.
The Evolution of Norms
Incomplete contracts rely on social norms to fill gaps. But where do norms come from? Game theory provides an answer through evolutionary models.
Imagine a population of people forming partnerships. Some write detailed contracts. Others rely on trust and norms. Which strategy survives?
If everyone writes detailed contracts, someone who builds a reputation for trustworthiness gains an advantage. They save legal costs while attracting partners who value flexibility. But if everyone relies purely on trust, opportunists who exploit that trust can profit.
The evolutionary equilibrium typically involves both strategies coexisting. Some situations call for contracts. Others run on trust. The population develops norms about which situations are which.
This explains why business culture matters so much. Some industries and regions developed high trust norms. Others remain adversarial. Neither is inherently superior. Each represents a stable equilibrium given that environment’s history and conditions.
Silicon Valley’s high trust culture emerges from repeated games among a tight community where reputation travels fast. Hollywood’s detailed contracts reflect an industry where projects involve one time teams and information asymmetries run deep. Both approaches make sense in context.
The Final Irony
The strongest argument for incomplete contracts comes from their honesty about limitations. A 50 page contract creates the illusion of certainty. It suggests all contingencies are covered. This illusion often proves dangerous.
When something not in the contract happens, parties feel betrayed. “We had an agreement!” The unwritten assumption was that the agreement was complete. The gap between expectation and reality breeds conflict.
An explicitly incomplete contract sets different expectations. Both sides acknowledge uncertainty upfront. They commit to working through unexpected developments together. When surprises arrive, nobody feels blindsided.
Game theory teaches that managing expectations is part of the game. Signals matter. Commitments matter. Creating realistic mental models in both parties leads to better outcomes than creating false certainty.
Three words on a handshake sometimes beat 50 pages of legal text. Not because details don’t matter. Not because trust alone solves everything. But because the right amount of incompleteness acknowledges reality, preserves flexibility, and builds the relationship capital that makes cooperation sustainable.
The future will surprise us. Game theory suggests we plan for that surprise by investing in relationships rather than pretending we can specify everything in advance. Trust isn’t naive. Done strategically, with clear eyes about when it works and when it doesn’t, trust is simply better game theory.
