profits—when reporting things to your employees. The four programmers are each going to think they’re very highly valued, because they’re making more than the average. Your poor receptionist won’t be so happy, but she no doubt knew already that the programmers make more than she does.
Now suppose you are feeling overworked and want to persuade your two partners, who don’t know much about critical thinking, that you need to hire more employees. You could do what many companies do, and report the “profits per employee” by dividing the $210,000 profit among the five employees:
Average salary of employees: $66,000
Average salary of owners: $100,000
Annual profits per employee: $42,000
Now you can claim that 64 percent of the salaries you pay to employees (42,000/66,000) comes back to you in profits, meaning you end up only having to pay 36 percent of their salaries after all those profits roll in. Of course, there is nothing in these figures to suggest that adding an employee will increase the profits—your profits may not be at all a function of how many employees there are—but for someone who is not thinking critically, this sounds like a compelling reason to hire more employees.
Finally, what if you want to claim that you are an unusually just and fair employer and that the difference between what you take in profits and what your employees earn is actually quite reasonable? Take the $210,000 in profits and distribute $150,000 of it as salary bonuses to you and your partners, saving the other $60,000 to report as “profits.” This time, compute the average salary but include you and your partners in it with the salary bonuses.
Average salary: $97,500
Average profit of owners: $20,000
Now for some real fun:
Total salary costs plus bonuses: $840,000
Salaries: $780,000
Profits: $60,000
That looks quite reasonable now, doesn’t it? Of the $840,000 available for salaries and profits, only $60,000 or 7 percent went into owners’ profits. Your employees will think you above reproach—who would begrudge a company owner from taking 7 percent? And it’s actually not even that high—the 7 percent is divided among thethree company owners to 2.3 percent each. Hardly worth complaining about!
You can do even better than this. Suppose in your first year of operation, you had only part-time employees, earning $40,000 per year. By year two, you had only full-time employees, earning the $66,000 mentioned above. You can honestly claim that average employee earnings went up 65 percent. What a great employer you are! But here you are glossing over the fact that you are comparing part-time with full-time. You would not be the first: U.S. Steel did it back in the 1940s.
• • •
In criminal trials, the way the information is presented—the framing—profoundly affects jurors’ conclusions about guilt.Although they are mathematically equivalent, testifying that “the probability the suspect would match the blood drops if he were not their source is only 0.1 percent” (one in a thousand) turns out to be far more persuasive than saying “one in a thousand people in Houston would also match the blood drops.”
Averages are often used to express outcomes, such as “one in X marriages ends in divorce.” But that doesn’t mean that statistic will apply on your street, in your bridge club, or to anyone you know. It might or might not—it’s a nationwide average, and there might be certain vulnerability factors that help to predict who will and who will not divorce.
Similarly, you may read that one out of every five children born is Chinese. You note that the Swedish family down the street already has four children and the mother is expecting another child. This does not mean she’s about to give birth to a Chinese baby—the oneout of five children is on average, across all births in the world, not the births restricted to a particular house or particular neighborhood or even particular country.
Be