Often when we face uncertainty, we would like to know more than simply the outcome; it would be nice to control the outcome and bring about the best possible result! A king might consult his wizard as well as his clairvoyant, asking the wizard to cast a spell to cause the desired outcome to occur. How much should such a wizard be paid for these services? The expected value of his wizardry naturally depends on the decision problem at hand, just as the expected value of information does. But the way to calculate the \"expected value of wizardry\" (to use Ron Howard's term) is very similar to solving the calculations for the expected value of perfect information. To demonstrate this idea, let's look again at Hugh Liedtke's decision situation as diagrammed in FIGURE 4.2 Hugh Uedtkes decision tree with chances(probabH- ieslinCMded. Counteroffer $5 Billion Texaco $3 Billion (0.33] Accept $2 Billion Texaco Refuses Counlerolfer Counteroffers Figure 4.2. Now consider the entirely hypothetical possibility that Liedtke could pay someone to influence Texaco's CEO, James Kinnear. Duesons 1. What would be the most desirable outcome from the \"Texaco Reaction\" chance node? 2. Construct the decision tree now with three alternatives: \"Accept $2 Billion,\" \"Counteroffer $5 Billion,\" and \"Counteroffer $5 Billion and Influence Kinnear.\" 3. Solve your decision tree from Question 2. What is the maximum amount that Liedtke could pay in order to influence Kinnear? Settlement Amount (5 Billion} 2 Texaco Accepts $5 Billion 5 HM?) 10.3 Final Court Decision D 10.3 Final Court Decision ' 0 Accepl $3 Billion 3