Link My iMpact  
Link Strategic Positioning Tool Kit  
To Executive Education
To Kresge Library

When Harry Bet Sally: Peer Effects at the Casino

1/15/2013 --

Study shows casinos can better target promotions by considering how friends influence fellow gamblers.

ANN ARBOR, Mich. — Underneath the cacophony of a slot machine's beeps and dings is an order of sorts, thanks to loyalty cards. Casinos know who gambles when and how much.

But people often go to the casino with family and friends, and the effects they have on each other's gambling habits are less obvious. Ross Marketing Professor Puneet Manchanda dove into two years of slot machine customer data from a casino and found that some promotions can actually cause a recipientís peer to gamble less.

Another negative effect he and co-author Hee Mok Park found occurs when a gambler's companion joins him or her, but doesn't gamble.

Manchanda and Park, a Ross PhD student, found these two new effects in addition to the better-known endogenous peer effect, which shows that two people consuming (or gambling) together tend to spend more. Their paper, "When Harry Bet Sally: An Empirical Analysis of Peer Effects in Casino Gambling Behavior," is one of the first to provide a holistic picture of peer effects at a casino.

"This is of obvious interest to casinos in terms of how they do their promotions," Manchanda says. "Our estimates show that there's a significant upside to getting a complete picture of interactions and adjusting promotions. We find that not only is there no spillover effect from a promotion — it actually has a negative peer effect."

Despite the size of the casino industry — it has greater revenue than sports teams and amusement parks — there's a paucity of research on the subject. Manchanda and his co-author also suspected that other peer effects were at work besides the endogenous effect. Slot machine data proved to be a rich source of information, since the vast majority of gamblers play slots and most use a loyalty card, which allows patterns to emerge.

The authors found that when one person receives a random promotion — not to be confused with a "comp" that rewards spending — and the other doesn't, the peer who doesn't receive the promotion spends significantly less. They call this the exogenous peer effect.

This suggests casinos experiment with less public ways to deliver random promotions to customers, such as though a smartphone or email.

"You canít eliminate the fact that one person might tell another, but it's worth trying to do these in a less obvious way," Manchanda says.

The study also shows that the negative exogenous effect varies within pairs of gambling partners. This suggests that one customer is more influential than the other. Casino managers can use data from the loyalty cards to target promotions to the more influential of the two.

The other negative effect could be harder to combat. When someone joins a friend or family member at the casino but isn't gambling, that friend or family member spends less than when he or she comes alone. This is the peer presence effect. Manchanda says casinos need to think about ways to encourage play from these customers.

As casinos continue to pop up in more U.S. states, and in more countries worldwide, studying consumer habits in this setting becomes more important, Manchanda says.

"This kind of research adds to the study of peer effects on consumers, but more specifically I think it has value as more casino revenue comes from Asia," he says. "Gambling is much more of a group activity in Asia, so understanding peer effects is very important in that context."

— Terry Kosdrosky

For more information, contact:
Terry Kosdrosky, (734) 936-2502,