1 Stanley Seiden Stanley Yu ECON 488 November 30, 2009 Tipping Behavior: An Experimental Approach

2 Question Addressed How does tipping behavior change based on the information presented on the bill and customers' method of payment?

3 Inspiration Irrational behaviors  Debit or credit card  round the total (including tip) to a whole number, even though there is no logical reasoning Suggested tips  Do people follow (strictly vs. as a guide)?  Convenient or insulting?

4 Background Each year in the United States, diners leave some $42 billion in tips at full-service restaurants Restaurants are now printing a “suggested gratuity” right at the bottom of the check. However, if you take a close look, you’ll see that restaurants are trying to bump up the tip by calculating the suggestion on the check total, not the total minus tax.

5 Hypotheses Bills with suggested tip amounts given will garner the largest tip amount across the three conditions  Guilt  Estimate based on post-tax suggested gratuity For given price, longer receipt (more items) might yield higher tips  Waiter is “doing” more Paying with a card, people would be more likely to irrationally tip so the total amount comes to a whole dollar

6 Results 13/22 participants (59%) tipped so total rounded to a whole number 8/22 participants (36%) used at least one suggested tip amount as given

7 Data 1

8 Data 2 8

9 Data 4 9

10 Data 5 10

11 Variance Observations 11 Highest with cheap meals Lowest with medium service

12 Analysis Results of expensive bill comparison deny hypothesis Effects of suggested tip amount:  Slightly higher tipping, but not by much  Slightly more predictable tipping amount Various irrational strategies common:  High tipping with cheap bills  Rounding  Tipping in general lower than suggested tip amounts 12

13 Takeaways While this seems minor, it probably is. However, people should be aware that the suggested tip is not accurate if they are depending on it to tip appropriately In small restaurants, intimacy is more important. Something as impersonal as a computerized “Tip Table” — especially a post-tax one — doesn’t foster that

14 Limitations of Study Limited sample size (22) Difficult to recreate complete dining experience  Many factors influence tip Participants are not handing over their own money There is the possibility that there is no consistent pattern to tipping in the real world!