University:
Stanford University
Course:
ECON 202 | Microeconomics I
Academic year:

2023
Views:

373

Pages:

2
Author:

T. Elton

Competitive Price Plans and Decision Trees Oliver (1986) mentions the use of decision trees in assessing brands in an established market, particularly an oligopolistic one in which competitors' responses can be anticipated with a reasonable degree of probability. He prefaces his example with some useful remarks on competitive price plans, distinguishing between initiatory and response strategies. The former includes precedence (being the first to take a particular action) and pre-emption (making a 'shut-out' move to exclude the competition). Precedence is used to catch rivals off guard, and make additional profits before retaliatory action by competitors; pre-emption is intended to make it difficult for rivals to follow a move, usually by cutting a price so savagely that competitors feel unable to match it. Response strategies comprise imitation and exploitation. An imitative response would be to match the first move, or 'better' it in kind - meeting a competitor's price cut with an even larger reduction, for example. An exploitive response would be one that contrasts with the original strategy, typically cutting prices in response to an increase by competitors. Obviously, the number of variables inherent in these different competitive stances lends itself to their examination in the form of a decision tree, in which the payoff can be expressed in terms of profits from expected sales under each alternative. For the sake of simplicity, only one competitor is assumed to exist (or all competitors act as one) and only two alternative prices are considered: 50p and 40p. If we choose an initial price of 50p, our competitor can respond with either 50p or 40p. The probability estimates that we assign to each response show that there is a 75 per cent chance that he will choose 50p, in which case we can meet his response by keeping to our original price of 50p or going down to 40p. Lowering the price to 40p gives us a 75 per cent probability of penetrating the market, in which case we shall make a profit of £80,000. We assign a 25 per cent probability to the alternative scenario that we will not penetrate the market at this price, in which case we would expect a much reduced profit of £40,000. From the table in Figure 11.12 it can be seen that eight alternative strategies are possible, and an expected value of the payoff associated with each strategy can be calculated from the decision tree. The expected payoff of strategy 8 setting an initial price of 40p and sticking to that price regardless of our competitor's response - is highest. Needless to say, the usual caveats regarding expected values should be considered before a statistical technique such as this is employed in decision making. A more complex example, together with a discussion of the Bayesian theory underlying the model, can be found in Green (1967).