Desktop version

Home arrow Sociology

  • Increase font
  • Decrease font

<<   CONTENTS   >>


When negotiating business deals, negotiators often face difficult choices that depend on their evaluation of risk. Consider, for example, this dilemma faced by your company in deciding to acquire either Company A or Company B.11 Company A has a $21 million value while Company B has a $15 million value. If the purchase price is the same for both companies, the decision to acquire Company A would be easy.

However, two risks arise if you acquire Company A. First, there is a 90 percent chance that the government will challenge the acquisition. Second, there is a 60 percent chance that the government will win. If the government wins, the value of Company A drops to $14 million. Even if the government loses, the value drops to $19 million because of legal costs incurred in fighting the government. The government will not challenge the acquisition of Company B.

Assuming that you will base your decision solely on the valuation of the two companies after factoring in risk, would you decide to acquire Company A or B? Most decision makers faced with this dilemma intuitively choose B. The decision tree is a tool that can help you determine whether this is a good decision.

A decision tree looks like a tree on its side. Decisions are represented by boxes. As depicted in Figure 4.2, the decision in this case is whether to acquire Company A or B. uncertainties are shown as circles. The two uncertainties in this situation are: (1) Will the government challenge the acquisition, and (2) if there is a challenge, will the government win?

After drawing the tree, probabilities are added (that is, 90 percent chance that the government will challenge the acquisition and 60 percent chance that the government [1]

Using a decision tree to calculate the impact of risk

Figure 4.2 Using a decision tree to calculate the impact of risk

will win). Endpoint values are also inserted: the valuation of Company A under three scenarios and the valuation of Company B. Finally an expected value of Company A is determined by calculating weighted averages: 60 percent of $14 million plus 40 percent of $19 million is $16 million, and 90 percent of $16 million plus 10 percent of $21 million is $16.5 million. Thus, the expected value of acquiring Company A ($16.5 million) is higher than the value of Company B ($15 million) and, if the decision is made on this basis alone, you should acquire Company A.

Decision trees can also be used to make other contract risk-related decisions. even simple "back-of-the-envelope” calculations using decision trees can make the available choices visible and easier to prioritize.

For instance, in the example above, Tim Cummins of IACCM described Microsoft's decision whether to insist on an indemnity clause in its contracts. As he noted, the indemnity clause brought minimal benefits in contrast to costs that included resource expenses (lawyer and management time) and lost cash flow (caused by delayed sales during the additional two to three months of contract negotiation).

Decision trees are useful in visualizing and testing decisions like the one that Microsoft faced. Let's assume that the contract clause in question provided Microsoft with $20 million in indemnity and that there is a 1 percent chance that it will lose $20 million and invoke the clause. (this probability should be easy to estimate based on past experience. In practice, the chance that such a clause would be invoked is probably less than 1 percent.) Let's also assume that the resource costs and cash flow costs to obtain the indemnity are $1 million. in effect, Microsoft would pay $1 million for the equivalent of a $20 million insurance policy. Given these assumptions, should Microsoft pay $1 million for this "insurance”?

the decision tree in Figure 4.3 depicts the 1 percent chance that Microsoft will lose $20 million if it drops the indemnification clause demand and the 99 percent chance that it will lose nothing. This results in an expected value of -$200,000 (0.99 x 0 plus 0.01 x $2 million). Based on these assumed values and probabilities (and not factoring in its attitude toward risk), Microsoft made a good decision when it dropped its demand for an indemnification clause, because the $1 million cost is $800,000 higher than the -$200,000 expected value of dropping the clause.

Using a decision tree to make an indemnification clause decision

Figure 4.3 Using a decision tree to make an indemnification clause decision

  • [1] this example is described in Victor, M.B. (1978) Predicting the cost of litigation. Strategy & Leadership, 6(6), 15-18.
<<   CONTENTS   >>

Related topics