Coordination Problems
Coordination Problems: Multiple equilibria exist β any of them would be fine, but only if everyone chooses the same one. Driving on the left and driving on the right are both viable; driving on different sides isn't. The challenge isn't choosing the best option but coordinating on the same option. Solved by conventions, communication, authority, and Schelling's focal points.
What Is a Coordination Problem?β
Coordination Problems arise when an interaction has multiple Nash Equilibria and the challenge is coordinating on the same one. Unlike the Prisoner's Dilemma (where individual incentives undermine the best collective outcome), coordination games have a cooperative character: everyone benefits from any equilibrium, as long as they all coordinate. The problem is not motivation but coordination itself.
Thomas Schelling's 1960 book The Strategy of Conflict formalised coordination games and introduced the concept of "focal points" β solutions that stand out by virtue of salience, convention, or common knowledge, and which people naturally gravitate toward without explicit communication. Schelling asked his students: "If you had to meet a stranger in New York City tomorrow, with no prior agreement, where and when would you go?" Most said Grand Central Station at noon. Not the only solution β but a focal one.
How It Worksβ
Coordination game structure:
β Multiple equilibria, all better than no coordination
β Individual benefit from coordinating (not conflicting incentives)
β Challenge: which equilibrium to coordinate on?
Examples of multiple equilibria:
β Drive left or drive right? (Both fine; neither is wrong)
β Meet at noon or 2pm? (Both fine if both agree)
β Use imperial or metric? (Both fine with coordination)
β Adopt standard A or standard B? (Both fine with adoption)
Coordination mechanisms:
1. FOCAL POINTS (Schelling points): salient solutions that
stand out without communication (noon, round numbers,
obvious landmarks, conventional norms)
2. COMMUNICATION: explicit agreement before choosing
3. AUTHORITY: designated decision-maker chooses
4. CONVENTION: historically established norms that
create shared expectations
5. LEADERSHIP: one party signals strongly, others follow
Three Real-World Examplesβ
Driving Side Conventionsβ
Driving on the left (UK, Japan, Australia) and driving on the right (US, continental Europe, China) are both pure coordination equilibria β neither is physically superior. The convention is determined by history. Switching is extraordinarily costly (requires changing all road infrastructure, vehicles, habits) once a convention is established. This illustrates that coordination equilibria are often path-dependent: the equilibrium reached is determined by historical accident, not by the equilibrium's intrinsic quality.
Technology Standards Warsβ
VHS vs. Betamax (1970sβ80s), Blu-ray vs. HD DVD (2000s), and USB-C vs. proprietary charging connectors are coordination games with network effects. The intrinsic quality of the format matters less than which format achieves coordination β both standards are better for all users than incompatibility. Network effects amplify small early advantages: the format with more content and devices attracts more users, which attracts more content and devices, which produces the coordination equilibrium. The "winner" is not necessarily the best format but the one that achieved coordination first.
Meeting Time and Social Conventionsβ
Why do business lunches default to noon, meetings start "on the hour," and parties begin at 7 or 8pm? These are focal points β there's nothing special about noon vs. 12:15, but noon is round and conventional, which makes it more salient as a coordination point. If everyone expects meetings to start on the hour, starting on the hour is rational regardless of whether the hour boundary has any inherent advantage. Conventions become self-fulfilling: because most people expect them, deviating is costly.
When to Apply Itβ
β Coordination analysis is valuable when:
- Multiple reasonable options exist and the key challenge is getting everyone to choose the same one
- Standard-setting in technology, law, or industry practice
- Designing conventions, defaults, or norms that others will rely on
- Solving "bootstrap" problems where value depends on participation
β It's not a coordination problem when:
- There's a clearly dominant option (no need for coordination, just calculation)
- The conflict is about who gets the value, not which equilibrium to choose
- Individual preferences for specific equilibria are strong and conflicting (closer to a Prisoner's Dilemma)
| Pairs well with | Why |
|---|---|
| Nash Equilibrium | Coordination games have multiple Nash Equilibria β the challenge is selection |
| Network Effects | Network effects amplify coordination advantage and make conventions sticky |
| Path Dependency | Coordination equilibria are path-dependent β history determines which equilibrium is reached |
| Social Proof | Social proof helps solve coordination problems by revealing what others are choosing |
Common Misuses and Limitationsβ
Treating all difficult collective decisions as coordination problems. Many conflicts are about who captures value, not which equilibrium to coordinate on. Labour-management negotiations, international trade disputes, and competitive market decisions have conflicting rather than aligned interests. Coordination analysis doesn't apply to genuine conflicts of interest.
Assuming focal points are sufficient. Focal points work when they're genuinely salient and shared. In global, multicultural contexts, what's salient in one culture may not be in another β reducing the power of purely cultural focal points.
Related Modelsβ
| Model | Relationship |
|---|---|
| Nash Equilibrium | Multiple Nash Equilibria are the defining feature of coordination games |
| Network Effects | Network effects make coordination advantages self-reinforcing |
| Path Dependency | Historical accidents often determine which coordination equilibrium persists |
Frequently Asked Questionsβ
What is a "Schelling point" and how does it work?
A Schelling point (or focal point) is a solution that people tend to choose in coordination problems without explicit communication β because it stands out by salience, convention, or common knowledge. Schelling demonstrated that people can often coordinate on one solution among many without communication, simply because certain options are more "prominent" or "natural." Round numbers, historical precedents, geographic landmarks, and established conventions all serve as focal points. They work because everyone knows that everyone else knows them β the solution's salience is common knowledge.
How do technology standards get established?
Through three primary mechanisms: (1) market-led coordination β one standard achieves early dominance through network effects and tips the market (VHS over Betamax); (2) formal standards bodies β industry consortia (IEEE, ISO, W3C) negotiate standards before products ship, preventing compatibility fragmentation; (3) government mandate β regulatory authorities designate standards (electrical socket specifications, measurement systems). Each mechanism has different speed, inclusiveness, and quality trade-offs. Market-led standards are fastest but may lock in inferior options; standards body processes are slower but produce broader buy-in.
What happens when coordination fails?
The cost of coordination failure depends on whether the alternatives are merely different (drive on left vs. right β both fine, but mixing is catastrophic) or one is genuinely better (metric vs. imperial β metric is more coherent). In technology, coordination failure produces fragmentation that reduces the value of the technology for everyone. In social norms, coordination failure produces ambiguity that increases transaction costs and conflict. The most costly coordination failures occur where incompatibility creates direct negative externalities.
Further Readingβ
- Schelling, T. (1960). The Strategy of Conflict β the foundational treatment of focal points and coordination
- Lewis, D. (1969). Convention β the philosophical analysis of coordination and convention
- Farrell, J. & Saloner, G. (1985). "Standardization, Compatibility, and Innovation." RAND Journal of Economics
Apply with AIβ
π Solve your coordination problem with MindMax β
This page is part of the MindMax Mental Models Knowledge Base.