Why is the Marginal Cost of Coffee Positive and the Marginal Cost of Cream Zero at Starbucks

I am at a Starbucks in Sioux City, Ia. using their wireless connection.  Thank god for Starbucks.  I get coffee and wireless access when I need it. 

As always, I see economics everywhere I go.  Here at Starbucks, as in many coffee shops, customers put their own cream into their coffee but the access to coffee is controlled.  Why does Starbucks give customers unlimited cream but charge them for extra coffee?  Are people more altruistic when it comes to using cream and sugar than they are when it comes to drinking coffee? 

Of course not.  The marginal utility of extra cream is near zero once you've filled your cup while the marginal utility from more coffee is greater than zero.  For me, the marginal utility of cream in my coffee is zero to begin with.  If there were unlimited coffee here, I'd probably depart more wired than the local electric company. 

Comments

[kb]

Um, I think dU/d(cream) goes negative at relatively small quantities of cream, but dU/d(coffee) stays positive over much wider ranges. You don't actually MIND being wired like an electric company, do you?

[spencer]

You also have to look at the cost of having an employee pouring the coffee.

It would probably cost more to have an employee pour the cream then the cost of
any extra cream customers take.

Compare it to the self-serfice soft drink dispensers at many fast food places.
Customers are free to refill their container -- and many do just before they leave -- but having the customer do the work is still cheaper then having an employee doing it.