9781118041581

Other Applications 551

From the table, we easily calculate that Pr(WƒG) .1/.2  .5 and Pr(WƒB) 

.3/.8  .375. After seeing a good test, the partners drill and attain an expected

profit of 200,000. After a bad test, what expected profit would drilling bring?

The requisite calculation is

Since drilling is profitable, the partners should drill even in light of a bad test result.

What is the overall expected profit with the test? After good tests (20 per-

cent of the time), the partners drill and earn $200,000. After bad tests (the

other 80 percent of the time), they also drill and earn $100,000. Thus, their

expected profit is (.2)(200,000) (.8)(100,000) $120,000. But this is exactly

the same profit they earn without the test. After a little thought, this should

not be a surprise. Without the test, the optimal action is to drill. With the test,

the optimal action is to drill. Since they take the same actions with or without

the test, they earn the same expected profit in each instance.

What is the general lesson to be learned from this example? Acquiring new

information is beneficial if and only if it has the potential to affect the man-

ager’s actual decisions. If it does not, the information is of no value.^2

OTHER APPLICATIONS

For pedagogical purposes, we have made intensive use of the oil-drilling exam-

ple. However, it is important to stress the generalapplication of information

issues in all types of business and public policy decisions. In a host of settings,

the decision maker is confronted with the task of quantifying his or her uncer-

tainty, that is, estimating a probability. Here are some examples:

• The largest consumer-products firms launch between 15 and 25 new
  products each year from a potential pool of 50 to 100 candidates. How

E()(.375)(600,000)(.625)(200,000)$100,000.

Wet (W) Dry (D) Total

Good (G). 1 .1. 2

Bad (B) .3 .5 .8

Total .4 .6 1.0

(^2) This point holds regardless of the decision maker’s attitude toward risk. Given the opportunity to
acquire information, a risk-averse manager solves the same decision tree as his or her risk-neutral
counterpart but uses expected utility as a guide. Information is valuable if the expected utility with
the information (after accounting for its cost) exceeds the expected utility without it.
c13TheValueofInformation.qxd 9/26/11 11:02 AM Page 551