Since this topic has generated a lot of interest, i think it is worth examining a little more. Let me get into the motivation behind tipping. I start with the premise that tipping improves economic efficiency.
Think about an individual trying to purchase a good. Most stores allow you to return the goods purchased within a time frame, therefore reducing the need for any form of customer "monitoring" prior to purchase. You can always buy the good and if it is of inferior quality, return it.Now think about the service industry. Can you give back services rendered by your dentist, if you feel he has not done a good job? Obviously not. This is where reputation acts as a medium for contracting. In the case of firms, reputation in terms of service quality is important for current and future consumption, especially because poor service tends to have a contagion effect on patronage.
Now given the firms position, the proxy for service quality is generated by the person attending to the customer. The person attending to the customer can either put in high effort or low effort (for ease of exposition, let us just look at the two polar cases). If he puts in low effort, it directly harms the reputation of the firm. Hence incentivizing the worker to choose high effort is imperative, from the firms perspective.Therefore the question now arises as to what tools are available to the firm to induce high effort from the worker? One obvious tool is monitoring the activities of each individual. If workers are found shirking and not delivering the high effort outcome, they can be punished, therefore directly influencing their behavior. Obviously this is costly, which implies that it has a proportional effect on service cost which naturally leads to pricing distortions, which again has a negative effect on patronage.
Given that the firm might find it costly to monitor and there are no indirect or direct incentives for the worker to provide high effort, the likelihood of the customer being hit by bad service is quite high. Therefore what are the options staring at the consumer? One potential solution is go to court. Litigation risk is an important factor modeled into the risk of any firm. Consumers have the right for quality service and can go to court to ensure that they are satisfied. But would you go to court for poor service in a restaurant? Again the expected costs will exceed the benefits. Another alternative is explicit contracting between the consumer and the worker. That would again be expensive and difficult to enforce. Given that there is no control over the number of consumers, the number of such contracts that each worker has to enter into will be numerous and hence this system will be inefficient.
Given this situation, in equilibrium, it is optimal for the buyer to pay even if he expects bad service. In the absence of any tipping practice, the price of the product will include a premium for service quality. Therefore regardless of service quality the consumer ends up paying the full price. Essentially the market for service quality just collapses, very much in the same way as the market for lemons (used cars, Akerlof's nobel prize winning work).
Suppose we introduce tipping into the model. The proposition is that, instead of inherently pricing service quality, let that be a random variable tied to service quality. In the setting mentioned about it can play the role of providing a good equlibrium. Suppose we make tipping discretionary on the part of the consumer, which is contingent on service quality. The burden of proof (from a legal setting) no longer is applicable. All the consumer has to do is reduce the tip or pay no tip at all. Hence the consumer ends up paying just for the product and not for service quality. The firm gets back its money for resources spent and does not have to spend any additional resources on monitoring or paying the waiter for service quality. In effect, the consumer and the firm are better off. In this situation, the worker is automatically incentivized to provide quality service or face the likelihood of not receiving any tips, thereby lowering his utility.
The role of the firm needs to be stressed in this case. The firm while setting the price can now remove the service quality aspect from the pricing, therefore reducing the overall price. Hence this saves any ex-post negotiation, which can prove to be costly. For instance consider the case of a restaurant. Assume that the service is poor. Then one option is just pay and keep quiet. However if you want to ensure fairness, then there should be a discount on the menu price for poor service quality. The question is how much should the discount be. This becomes a matter of negotiation between the firm and the consumer and the potential solutions are not bounded. However if service quality is not priced and it is purely discretionary in the form of tips, the firm is not affected and neither is the consumer. The consumer still pays for the food and can choose the magnitude of tipping based on the quality of the service. Now note that in such a situation the equilibrium can be distorted by the consumer because, regardless of the service, he can choose to maximize his utility by not tipping.
Now comes the issue of tipping limits. Given that there are consumers who dont tip regardless of the service, the limit of 15% is kind of like a social norm to force consumers not to play that equilibrium move. Another potential thought could be like insurance. While several people will adhere to the 15% norm, some will not. Therefore the ones that do subsidize the ones that dont.
In effect, it serves more like a psychological tool trying to persuade consumers not to misuse the fact that they can get away without paying for service quality.
Sunday, January 28, 2007
Tipping - My take
Posted by Mad Max at 11:18 PM
Subscribe to:
Post Comments (Atom)
Posts
6 comments:
But that still leaves Mindframes question unanswered doesn't it? Why should I pay on price of food ordered- considering that in many typical day to day menu fares the effort to serve an expensive dish is equal to (more or less) to serving simple fare. In most cases in typical restaurants (remove jacket/tie places) its the ingredient price that rougly (very roughly) translates to the menu price-- so why make that the basis for target percentage?
Wouldn't it be more optimal to setup a simpler system that takes into account actual effort of serving. Some examples:
1. Number of people being served
2. Rough number of requests catered to
3. perhaps number of dishes served
Have each contribute some points and then translate points to $$$.
oops forgot most important point:
4. Perceived quality of service
@ Manohar: I agree. First we have to note that there is no contract which enforces you to pay 15%. It is only a reference. And the reference point is in terms of price, because that is the easiest to quantify. Further in most sit down restaurants the price of an entree tends to be fairly homogenous. for example, $ 15.99 might be the price of say 10 dishes. Therefore from the pricing one perceives that the utility to the customer is relatively same regardless of the dish chosen. While I agree that the assumption is stiff, it is a directly observable measure which is not random. Hence it is easier to have a reference point in terms of menu price.
From the other options mentioned. 1, 2 and 3 is just a scaling factor. For instance if 1 person eats a dish worth 15 dollars, the likelihood that 4 persons will eat something worth 60 dollars is quite high. Therefore using any metric one can still work backwords to ensure a 15% tip.
The last point is ambigous. As noted in the blog, it provides opportunity for consumers to play a bad move and hence will not serve the purpose in the set up that i have talked about.
what say u?
Very well written mad-max... I think you talk about how to ensure service quality in a restaurant and how tipping plays a role to make both sides happy... You also say that percentage of bill is a good measure of service offered, which is highly debatable. But, I cant think of a simple alternative, though the suggested alternatives from Mano makes more sense (may be difficult to implement). Also, the term "service" is a highly debatable and an ambiguous term too. To me, if I get what I want from the menu without having to wait for a while, it covers most part of "service". But then, I think that is what the waiters are meant for... I usually dont see any greater incremental service in most places, may be a good smile...but, will I be willing to pay 15% of my bill towards that extra smile, I wouldnt....
@ Mindframes: I do not explicity say that the percentage on the bill represents service quality.
How about this explanation. Suppose ex-ante every consumer expects good quality service on average. Given that firms know such expectations exist, they are placing a lower bound on what that expectation could be valued. NOTE THAT THERE IS NO EXPLICIT OR IMPLICIT CONTRACT THAT THE TIP HAS TO BE 15%. It is only a reference point and one which is easier to contract on.
On the alternatives suggested by Mano, while i do agree that those metrics might be helpful, it will not be economically efficient to implement them. I will explain in detail in another blog.
mad-max: You da man! waiting for your further explanation in another blog. :)
Post a Comment