# Tip percentage declines as bills increase

In the world of gratuities, the bigger they are, the harder they fall.

Psychologists at Washington University in St. Louis have shown that the larger the bill, the smaller the tip percentage that food servers, hair stylists and cab drivers receive.

Leonard S. Green, Ph.D., Washington University professor of psychology in Arts & Sciences, and Joel Myerson, Ph.D., research professor of psychology, compiled data from nearly 1,000 tips left in restaurants, hair salons and with cab drivers.

Their findings indicate that the percent of the tip actually decreases with the amount of the bill across all three tipping situations. The researchers also found that with bills more than \$100, the percent of the tip levels off — if the bill is \$200, the server is likely to receive the same amount as if it were \$100.

In the study by Green and Myerson, servers at two restaurants, hair stylists at two hair salons, and two cab drivers from different companies kept data for the researchers.

“We’ve shown a magnitude effect in tipping,” said Green, who, with Myerson, has done numerous studies in behavioral economics. They examined tipping because it provides a good model for understanding economic and psychological decision-making.

Magnitude effects are economic phenomena that are contrary to standard microeconomic theory. The standard theory argues that the relative value of two amounts does not depend on the absolute amounts involved.

In contrast, a magnitude effect occurs when relative value is affected by absolute amount. For example, previous work by Green and Myerson has shown a magnitude effect in choices that involve immediate versus delayed amounts of money. Participants in a study were given a choice between two hypothetical dollar amounts:

“If you could have \$1,000 in one year, how much would you accept right now instead of having to wait the whole year? You might say, ‘Rather than wait, I’ll take \$500 just to have it now’,” Green said. “If you could have \$100,000 in a year, microeconomic theory would then predict that you would take \$50,000 now. In both cases, an amount delayed for one year loses 50 percent of its value.

“However, we have found that this prediction is incorrect. If you were to take \$500 now rather than \$1,000 in a year, you would demand more than \$50,000 now — for example, \$70,000 — if the alternative were \$100,000 in a year. Thus, with a smaller delayed amount (\$1,000), the relative value of an equivalent immediate amount was 50 percent, whereas with a larger delayed amount (\$100,000), the relative value is more than 50 percent.”

In a paper published in the current Psychonomic Bulletin & Review (Vol. 10, No. 2), Green and Myerson, working with an undergraduate, Rachel Schneider, have now shown a magnitude effect in tipping, as well.

For example, let’s say you had a bill of \$10 and left a tip of \$2, or 20 percent. The next week you’re at a trendy restaurant and dinner for two is \$80. You leave a tip of \$12, or 15 percent. The relative value of what is considered an appropriate tip has been influenced by the absolute amounts involved.

One little “tip” the researchers pass on to current food servers is: Encourage separate checks at large tables.

“The magnitude effect as seen in our study would suggest that, for a table of six, and a bill of \$25 a person, a single bill of \$150 would give the server a smaller percentage tip than six individual bills of \$25 each,” Green said. “For the server, it pays to give individual bills.”