Las Vegas casino card dealers make a considerable part of their income from tips (the "toke," which is supposedly short for token of appreciation). The customary practice is to pool all dealers tips and divide them up pro rata. According to the W$J, Steve Wynn has changed the practice at his eponymous casino:
In August, Steve Wynn, the CEO of Wynn Resorts Ltd., decreed that, in the case of dealers, the coveted tip pool will now be expanded to include about 220 of their supervisors. The move, the company says, is meant to correct a strange inequity in the casino workplace: Thanks to tips, dealers make thousands of dollars more than some of their bosses do. ...
Bringing supervisors into the tip pool means that dealers' income will drop, though it isn't clear how much. Wynn Resorts says the drop would be 10%, while dealers contend it will be closer to 15% or 20%. Managers would see their pay jump to about $95,000 annually with new tip income factored in, the company says.
The 64-year-old Mr. Wynn and his lieutenants have said the change was necessary to hire and retain casino supervisors. Andrew Pascal, Wynn Las Vegas's president and Mr. Wynn's nephew, said in an interview that there is little incentive for supervisors to stick to a management career track, "when they look at the people they're managing and realize how much more they're making."
From an economic perspective, this change doesn't seem to make much sense.
Economists use the phrase “agency costs” when talking about the risk that employees will shirk or cheat. In economic lingo, a sole proprietor with no employees—i.e., agents—will internalize all costs of shirking, because the proprietor’s optimal trade-off between labor and leisure is, by definition, the same as the firm’s optimal trade-off.
In everyday language, if you own a business, you’ll decide whether you’d rather make an extra buck by staying open longer or go home to watch American Idol. You’ll make the trade-off between work and leisure that makes you happiest.
Agents of a firm, however, do not internalize all of the costs of shirking: the principal reaps part of the value of hard work by the agent, but the agent receives all of the value of shirking. If I’m making $10 an hour, I’ll make $80 per day no matter how hard I work, so I’ll be tempted to work just hard enough to keep you from firing me. I’ll enjoy goofing when I’m supposed to be working, maybe by surfing the internet, while you make less money, which is what economists call shirking.
Accordingly, an essential economic function of management is monitoring employees—management meters the marginal productivity of each employee and then takes steps to reduce shirking.
Tipping is an efficient solution to the principal-agent problem in some service industries, as Eric Posner explains using restaurants as an example:
Waiters earn a lot of money from tips; at the same time they are paid a flat wage by the restaurant. This is exactly the sort of mixed contract that the agency model predicts. The flat wage protects the waiter from slow nights and miserly customers; the tips give the waiter an incentive to do a good job. But there is a twist to this story. Waiters often pool their tips and divide them evenly; other times, the managers require them to pool their tips and divide them evenly. Notice that this increases insurance against bad tippers, but reduces the incentive to provide good service. Part of the reason for pooling from the manager’s perspective comes from the multiple agent problem: when waiters depend heavily on tips, they have no incentive to cooperate with each other. This is why a waiter who does not serve your table will sometimes be rude or impatient when you ask him for some help. To enhance cooperation, the manager requires pooling of tips. And the waiters might not object. From their perspective, pooling allows them to insure against the chance that on a given shift all one’s customers are bad tippers.
Notice how including managers in the toke pool introduces at least two sources of inefficiency. First, as long as everyone in the pool is a waiter (or a casino dealer), the potential for reduced incentives resulting from pooling is minimized because: (1) pool members understand that things come out in the wash, some nights nights they won't benefit from pooling (they were at a $1000 table with a lavish tipper) but sometimes they'll benefit (they were at a $5 table), so they'll be willing to work hard anyway; (2) social norms and peer pressure within the pool participants discourage shirking and free riding.
Once managers are included in the toke pool, the insurance aspect of pooling is diminished. After all, presumably the managers are not interacting with customers and therefore are not generating tips. There is no benefit to the dealers from including the managers in the pool; only costs.
In addition, because managers and dealers are in a hierarchial relationship, including them in the toke pool makes social effort norms less potent. Norms are strongest among homogenous groups, so introducing heterogeneity tends to weaken them.
Finally, including middle management in the toke pool may change the incentives of those managers to monitor dealers. They may focus on maximizing the size of the toke pool instead of focusing on maximizing casino profit.
All in all, it looks like an economically unsound idea.
