Executive Summary: Researcher observes that servers get increased tips due to unchangeable or hard-to-change factors such as age, breast size, hair color, body mass, and race, as well as seemingly arbitrary changes such as squatting and drawing happy faces. He also observes that tipping’s connection to service quality is real, but weak.
I repeat, the system of tipping is bad; with the way the current system is set up, tipping is not bad. I tip between 18 and 20% when eating out. I may tip more than that percentage if I used a coupon. If I don’t think I’ll be able to afford the meal and the tip, I choose not to eat out.
That out of the way, let’s go hypothetical.
The hypothetical jerk boss
Suppose you wanted to start a business that would require hiring employees.
Suppose further that for whatever reason, you wanted your attractive female employees to get paid more than your unattractive female employees who do the same quality of work. Suppose also that you wanted your slender female employees to make more, your large-breasted female employees to make more, your female employees to see their income peak if they’re in their 30s, and your blond female employees to make more. Not only that, for some reason, you want all employees who are more prone to squat, but only female employees who draw happy faces to make more too. Oh, and just to round things out, let’s say you wanted your white employees to make more than your minority employees too.
Now, if you were this sort of discriminatory, racist, ageist, entrepreneurial weirdo, you might think it would be difficult to set up a system that would pay according to your crazy whims, but you’d be wrong. All of these discrepancies exist in service jobs where some income comes from tipping.
Michael Lynn, the Burton M. Sack ’61 Professor in Food and Beverage Management at the Cornell University School of Hotel Administration has been researching all aspects of tipping for decades and published over 35 academic papers on the topic.
In his experiments, Lynn generally seeks to isolate the effects of individual factors — for example the age of a server — and find the relationship between increasing or decreasing that factor and the level of tips. Here are some details he observed in his studies:
- Age: Among waitresses, percent tip reached its peak value among women 36 to 40 (1)
- Breast size: Percent tip increased as breast size increased, going up nearly 1/2 a percentage point with each additional cup size (1)
- Hair color: Blonde waitresses received tips over one percentage point greater than their non-blonde counterparts (1)
- Body mass: Body mass index (BMI) was found by entering height and weight of waitresses into a BMI table such as that available from the National Institutes of Health here; percent tip decreased by 1/10th a percentage point for each increase in BMI (1)
- Squatting: Waiters and waitresses saw their tips increase approximately 3% when they squatted in their initial interaction with customers, rather than stand (2)
- Drawing a happy face: Waitresses who drew a happy face on the back of the check received tips that were 5% higher than when they didn’t, whereas waiters saw no statistically significant difference when they drew happy faces (2)
- Race: Black servers received tips that were 3.25% less than white servers (3)
Servers have control over some of these factors, but not others. Age and race in particular cannot be targeted for discrimination, and so it’s tough to see why results like these are OK.
Tipping as either an incentive or reward for good service?
So the system of tipping allows for discrimination, but what about quality of service?
Quality of service is certainly one factor that servers have control over, which makes this an important question. Many customers will cite a desire to reward good servers as a reason that they tip, and restaurateurs may imagine the system useful because it enlists the help of their customers in ensuring quality control.
To answer this question, Professor Lynn used results from 13 studies and found the relationship between good service and increased tips to be real, but weak — certainly weaker than we might expect. He concludes with this advice to managers:
Given the small size of this relationship, restaurant managers should not rely on tips as the sole incentive for their employees to deliver good service. Nor should managers use server tip averages as the only measure of server performance. Tips are a means by which consumers reward good service, but (in restaurant settings at least) they may not be the management tool that economists and some restaurant managers believe them to be.
Does anyone have thoughts on this topic?
- Lynn, Michael (2009). “Determinants and consequences of female attractiveness and sexiness: Realistic tests with restaurant waitresses.” Archives of Sexual Behavior, 38, 737-745.
- Lynn, Michael (1996). “Seven ways to increase your servers’ tips.” Cornell Hotel and Restaurant Administration Quarterly, 37 (June), 24-29.
- Lynn, Michael, et. al. (2008). “Consumer racial discrimination in tipping: A replication and extension.” Journal of Applied Social Psychology, 38, 1045-1060.
- Lynn, Michael & McCall, Michael (2000). “Gratitude and gratuity: A meta-analysis of research on the service-tipping relationship.” Journal of Socio-Economics, 29, 203-214.