What the game theory lecture is really about.
The most durable insight in economics is often the one that destroys a false dichotomy. Ashley Hodgson's lecture does exactly this for competition and cooperation. The common framing treats them as a dial: more of one means less of the other. The game-theoretic view destroys this. In every game Hodgson examines, competition and cooperation are simultaneously active, operating at different levels of the same interaction. They are not a spectrum. They are a structure.
Two classic games carry the argument. The Battle of the Sexes shows that even a game with only cooperative motives (both players want to end up together) contains internal competition (over which equilibrium they end up in). The Prisoner's Dilemma shows the mirror image: a game defined by competitive incentives can sustain cooperative outcomes when the game is repeated. In both cases, the two forces don't cancel. They cohere.
Three kinds of edits to the latticework follow: models amplified, models bent, and new models that earn a place of their own.
Models the lecture amplifies.
Nash Equilibrium as a structural lens, not just a solution concept.
Nash equilibrium is usually taught as the endpoint of strategic analysis: find it, and you know the outcome. Hodgson uses it as a structural lens: the existence of multiple equilibria reveals the embedded competition. In the Battle of the Sexes, both equilibria are cooperative (both players end up at the same venue), but the players compete over which one. The equilibrium count tells you how much competitive pressure exists within the cooperative frame.
Repetition changes the dominant loop.
The Prisoner's Dilemma in one shot has a clear dominant loop: defect. Add repetition and the loop changes: the threat of future retaliation makes cooperation the rational strategy. Feedback loop theory says that the same system can have different dominant loops depending on conditions — time horizon being one of the most important. The PD is a textbook demonstration: the defection loop is dominant in the short run; the cooperation loop becomes dominant when the game is repeated.
The ancient proverb, formalized.
The Arab proverb — "me against my brother, me and my brother against my cousin, me my brother and my cousin against the world" — is a folk statement of a game-theoretic truth: at every scale, the competitive and cooperative forces reorganize around the level at which an external threat appears. Hodgson formalizes this: the in-group competes internally while cooperating externally, and the level at which the game is played determines which force is dominant at any moment.
Incentive design determines which force wins.
The lecture's implicit lesson about the Prisoner's Dilemma is that the outcome — cooperation or defection — depends on the incentive structure, not on the players' character. Change the time horizon, add a credible retaliation threat, or change the payoff matrix, and the same players will choose differently. This is the incentives model at its most precise: you do not need to change the players to change the outcome. You need to change the rules of the game.
The firm as a nested competitive-cooperative system.
Hodgson's corporate example is a clean instance of nested systems: departments compete internally, while the firm competes externally. At each level of the hierarchy, the same pattern repeats. The firm is not simply "cooperative" or "competitive" — it is cooperative at one level (against external competitors) while competitive at another (between internal departments). Any attempt to label an organization as one or the other is picking a level and ignoring the rest.
Models that don't survive intact.
Most real games are not zero-sum.
Zero-sum thinking assumes that one player's gain is another's loss. The Battle of the Sexes is the direct refutation: both equilibria are gains over the off-equilibrium (going to different venues). The cooperative structure creates a positive-sum set of possible outcomes, within which competition allocates the gains. Applying zero-sum thinking to a situation with this structure systematically misses the size of the cooperative gain being competed over. The risk: leaving cooperative surplus uncaptured by framing every interaction as purely extractive.
Rationality and cooperation are not opposites.
The naive reading of rational self-interest says agents defect in a Prisoner's Dilemma because defection dominates. The lecture shows that this is only true in a one-shot game. In the repeated game, rational self-interest — fully and correctly calculated — generates cooperation. The "rational actor defects" claim is an artifact of an incomplete model (single-period). It is not a claim about rationality itself. The corrected principle: rational self-interest, projected across the full time horizon including future interactions, often selects for cooperation.
Models worth adding to the latticework.
Fractal Game Nesting.
At every scale of social organization — individual, department, firm, industry, nation — the same pattern repeats: competition within the level, cooperation against the level above. This is not a metaphor but a structural property. Whenever you analyze a conflict or cooperation at one level, the correct question is: what is the level at which this group cooperates? And: at what level below is it competing internally? The proverb is the pattern. The game theory is the mechanism.
The Repeated Game Flip.
There is a class of games in which adding time (repetition) flips the dominant equilibrium from competitive to cooperative. The Prisoner's Dilemma is the canonical example, but the pattern is portable: any situation in which the one-shot dominant strategy is destructive should be tested for whether repetition — credible future interaction — flips it. The practical question: is this a one-shot or a repeated game? Getting this classification wrong is one of the most common strategic errors.
Pie Growth as the Cooperative Denominator.
Within a firm, departments compete for budget. But Hodgson notes that even purely self-interested departments want the organization to grow — because a larger pie means the manager can fund both departments rather than forcing a zero-sum choice. The implication for organizational design: cooperative behaviors can be sustained between competing units if the overall system's growth makes the competition less zero-sum. The cooperative denominator is not goodwill — it is expanding total resources.
The Coherence Test.
Before labeling a situation as "competitive" or "cooperative," run the coherence test: ask at how many levels each force is simultaneously active. If both forces appear at multiple levels simultaneously, the situation is a coherent mixture — and interventions aimed at one force without accounting for the other will misfire. The coherence test is a diagnostic tool: it tells you whether you are looking at a genuine trade-off or an embedded structure where both forces are load-bearing.
When to reach for which.
The latticework, re-woven.
The most useful thing this lecture does is replace a dial with a structure. Competition and cooperation are not a spectrum with more of one meaning less of the other. They are simultaneously active forces whose relative dominance depends on the level of analysis, the time horizon, and the payoff structure. Treating them as a trade-off — as most folk wisdom does — makes interventions less effective, because it addresses one force while ignoring the other.
Competition and cooperation are opposites that cohere — these are not opposites, they're actually very much enmeshed in the fabric of the world. — Ashley Hodgson
The latticework addition: when analyzing any strategic situation, do not ask "is this competitive or cooperative?" Ask "at which levels is competition dominant, at which levels is cooperation dominant, and what are the conditions — time horizon, incentive structure, pie size — that determine which force prevails?" That is a harder question. It is also the correct one.