Game Theory for Negotiation & Competition
In a prisoner's dilemma, individually rational choices produce a worse joint outcome. Price wars are a classic business example.
In repeated games, strategies like tit-for-tat — cooperate first, then mirror — sustain cooperation.
Credible commitments change the game: a public price-match guarantee can discourage competitors from cutting prices.
Map the zone of possible agreement (ZOPA): the overlap between each side's walk-away points.
The Prisoner's Dilemma is a canonical game theory scenario where two parties, acting in their own self-interest, result in a suboptimal collective outcome. The 'Nash Equilibrium' in a one-shot Prisoner's Dilemma is for both parties to defect (e.g., both cut prices), even though they would be better off if they both cooperated (e.g., maintained prices). This is because each player fears that if they cooperate while the other defects, they will receive the worst possible outcome, making defection the individually rational choice regardless of the other's move.
In repeated interactions, strategies like 'Tit-for-Tat' can foster cooperation. This strategy begins by cooperating on the first move and then simply mirrors the opponent's previous action. It is effective because it is 'nice' (never defects first), 'retaliatory' (punishes defection immediately), 'forgiving' (returns to cooperation after a single punishment), and 'clear' (its rules are easily understood by the opponent). However, its primary vulnerability is a susceptibility to signal errors; an accidental or misinterpreted defection can trigger a prolonged cycle of mutual retaliation.
You, the next time you negotiate a major contract, don't just focus on your walk-away price. Map the full ZOPA by considering non-price variables like delivery schedules, payment terms, or support levels that might create value for both sides.
Learning objective
Map incentives, alternatives and repeated interactions without encouraging unlawful coordination.
Understand the mechanism
A payoff matrix makes strategic dependence visible: your result depends on your action and another party's action. A Nash equilibrium is a set of choices where no player benefits by changing alone, given the others' choices. It need not maximize the group's outcome.
Repeated interactions can support trust when future consequences matter, but imitation or retaliation is not always optimal. Errors, asymmetric information and legal constraints change the game. Never coordinate prices or markets with competitors; competition law applies regardless of a theoretical model.
Put it to work
For negotiation, map reservation points and nonprice issues. A positive ZOPA means price agreement may be possible, not that a deal is guaranteed. Financing, diligence and execution conditions still matter.
Test whether commitments are credible and costly to reverse. A slogan is not a binding constraint. Analyze uncertainty about the other side rather than assuming their incentives match yours.
Worked example
- 1
Seller requires at least $8m; buyer can justify at most $10m. Price ZOPA is $8m–$10m.
- 2
If the buyer must spend $3m on repairs not yet in its model, the maximum could fall to $7m, eliminating that price overlap.
- 3
Different timing or scope might create value, but do not force an agreement that fails each side's real alternatives.
A model clarifies incentives; it does not remove unknowns or legal duties.
Your practice lab
Seller floor $12m; buyer ceiling $11m. Name a legitimate way to investigate whether a package could work without pretending there is a price ZOPA.
Personal scratchpad · not graded or saved
The Cloud Capacity Commitment
A dominant enterprise software firm, 'Titan Corp,' faced a rising competitor, 'Nimbus Cloud,' which was aggressively discounting its services to win market share. Titan's executive team knew that a tit-for-tat price war would erode the high margins their investors expected. Instead of just matching prices, they analyzed Nimbus's core constraint: limited server capacity, which made it difficult to onboard very large clients quickly.
Titan launched a new program called the 'Enterprise Onboarding Guarantee.' It was a public, contractual commitment to migrate any new client with over 10,000 employees to their platform within 90 days, or pay a penalty of $1M. This was a credible commitment because Titan had vast, underutilized cloud infrastructure, making the promise low-risk for them but nearly impossible for Nimbus to match without massive, costly investment. The commitment didn't focus on price, but on a non-price variable (speed and capacity) that was critical to large customers.
Titan changed the game not by playing the competitor's game (price), but by creating a credible commitment around its own unique strength, making the competitor's main strategy less effective.
Illustrative example — names and figures are composites for teaching.
Identify a key competitor for your venture. Describe one potential 'game' you might play with them (e.g., pricing, feature launches, talent acquisition). What credible commitment could you make to shift the game's outcome in your favor, without illegally coordinating?
Example: Our key competitor is a legacy consulting firm. The 'game' is winning key enterprise accounts. Our credible commitment is a 'Results-Based Guarantee': we contractually tie 50% of our fees to the client achieving specific, pre-agreed performance metrics within 12 months. This is credible because it puts our revenue at risk, and it's something they can't easily match due to their rigid, time-based billing structure.
Go deeper — outside resources
- Program on Negotiation at Harvard Law School · Harvard University
- Game Theory (Open Course) · Yale University
Key ideas
- Individual logic can trap groups
- Repetition enables cooperation
- Credible commitment shifts outcomes
- Find the ZOPA
Try it: Identify a repeated game in your industry and the cooperative strategy that wins it.
Quick check
1. Seller minimum $8M, buyer maximum $10M. The ZOPA is…
2. In game theory, what is the definition of a Nash Equilibrium?
3. Which of the following is a known vulnerability of the 'Tit-for-Tat' strategy?
4. Two software companies are the only providers in a niche market. Company A is considering an aggressive price cut to gain share. If this is modeled as a one-shot prisoner's dilemma, what is the most likely outcome?
5. Which of these is the strongest example of a credible commitment intended to alter a competitor's behavior?
6. A positive Zone of Possible Agreement (ZOPA) on price guarantees that:
7. Under competition law, which action is explicitly forbidden, regardless of game theory models?