Game Trees in Business Strategy
Why Studying Business Strategy? | Game Trees in Business Strategy | Bargaining | Notes
Game Trees
Two Kinds of Strategic Interation: Sequential & Simultaneous
The essence of a game of stratefy is the interdependence of the players' decisions. Theses interactions arise in two ways. The first is sequential; each player makes alterning moves. The players must look ahead to how his current actions will affect the future actions of others, and his own future actions in turn. The second is simultaneous; the players act at the smae time, not knowing what the others' current actions. When you play a strategic game, you must determine whther the interation is simultaneous or sequential.
Sequential-move games
Rule: Look Ahead & Reason Back
The general principle for sequential-move games is that each player should predict the other players' future responses, and use them in calculating his own best current move. Therefore, players should anticipate where their initial decisions will ultimately lead, and use this information to calculate the best choice. Most strateifc situations involve a long sequence of decisions with several alternatives at each. Games trees of the choice in the game gives a visul aid for successful application of the rule of looking ahead and reasoning back.
Consider a business example that has a game tree. Suppose the market for MP3 players in Canada is dominated by Apple's iPod, and a new firm, iRiver is deciding whether to enter this market. If iRiver enters, iPod has two choices: accommodate iRiver by accepting a lower market share, or fight a price war. If iPod accommodates the entry, each makes a profit of $1,000,000. But if iPod starts a price war, this causes iRiver to lose $2,000,000 and iPod to lose $1,000,000. If iriver does not enter the market, its profit is zero. Here is the game tree and the profit for each outcome.
What should iRiver do? The outcomes "accommodation" and "price war" as alternatives arse by chance. If probabilities of the two are thought equally likely, each gets a probability of 1/2. then calculate the average profit that iRiver can expect from entry by multiplying each profit or loss with the corresponding probability. IRiver gets (1/2)($1,000,000)-(1/2)($2,000,000)=-$500,000. Since this is a loss, you would recommend iRiver to keep away from Canada.
But how can we get the probability estimates? The probabilities come from iRiver's beliefs about iPod's profits in each of these cases by using game theory. In order to estimate what iPod will do, iRiver must first estimate iPod's profits in the different scenarios. Then the players look forward and reason backward to predict what the other side will do.
