Are the prisoner's dilemma and game theory relevant to this competitive workplace scenario?

Question

A friend of mine who works as a car sales person recently explained to me how she and her colleagues are paid. They get a very low base salary and make most of their money of off commissions and bonuses. They get 25% of the profit that the dealership makes on each car they sell, an additional 500$ if they sell more than 5 cars per month, a free car lease if they manage to maintain a steady rate of more than 10 cars per month (they have to pay for the lease if they fall below that rate of sales), and a bonus for the top seller at the dealership each month. They keep a score board in the office of how many each person sold per month.

The part that surprises me is that the system is setup such that all the sales people are in direct competition with each other. Although the dealership profits either way, from the sales person's point of view it is a zero sum game, an additional sale closed by a colleague is one less sale for me, assuming that the number of potential customers walking through the door is predetermined.

This isn't just theoretical, she has relayed all sorts of stories of back stabbing and selfish behavior among the team. Nor is it limited to her car dealership, it is the norm across dealerships, at least in our part of the US.

My philosophical question here is that this doesn't make sense to me from the dealership's point of view: Doesn't game theory in general indicate that in such situations, cooperation will benefit the group better than selfish behavior? Doesn't the dealership stand to benefit more from having the sales people work together to achieve more sales, than for each individual to try to sell as many cars as possible, even at the cost of preventing other people from closing sales?

Consider the following scenario: There are two sales people, one very experienced and the other a beginner. Two customers walk in at the same time, one is still shopping around and hasn't made up their mind, while the other is decided and is ready to sign a lease on the spot (In real life, they can tell these two types of customers apart from experience, and they call the second type spoons).

• They way things are set up now, the experienced person (who typically has more seniority in the office) will try to 'snatch' the spoon and leave the undecided customer to the rookie - since this allows them to maximize their income. Being inexperienced, the second sales person has a low probability of convincing the customer of purchasing a car. The dealership as a whole loses, since the experienced sales person would have had a higher chance of closing the deal. That's what usually happens in real life.

• If their pay structure was more inductive to cooperation, the experienced sales person would have handed the sure fire customer over to the rookie, and focused their efforts on the undecided customer. The dealership as whole would have potentially made two sales instead of one.

From a game theory point of view, isn't the second scenario more beneficial for the dealership? In particular isn't this scenario similar to the prisoners dilemma, where even though it is possible to maximize individual profit by being selfish, the group as a whole benefits more from cooperation than from selfishness?

To summarize:

• Is it correct to assume that game theory in general indicates that cooperation is better in such situations than competition?
• Is my reasoning w/r to the experienced sales person helping the rookie being a more beneficial strategy to the dealership as whole correct?
• Is the prisoner's dilemma indeed relevant to this situation or not?

Answer 1

These interactions can be better modeled with Drama Theory, which covers a set of individuals playing multiple games in a row, with the ability to change their decision making process between game theory games. This better models human interactions.

If you have drama theory, you can explore more complicated interactions, such as a subset of drama approaches known as tournament theory, where it is believed that people competing for a large share of the pie yields better results than sharing the pie. Tournament theory is one tool which has been used to describe why executive pay is so high: its not because the executive deserves it, but rather because it acts as an incentive for the subordinates to strive to one day be paid highly.

Also, consider a secondary issue which is not mentioned in your example. The dealership is not simply in the business of selling as many cars in the short term as possible. It also needs to ensure it has the best salespeople in the future. Determining who is a good sales person is not an easy task. In fact, it can be frighteningly difficult to build a metric for determining who is a good employee in general, and sales is even worse than the average case. The tournament structure has the added benefit of assisting management in identifying who fits their mold, and who doesn't. If this benefit is sufficient, it may overwhelm the management's ability to sell a few more cars in the short term by having everyone work together.

Answer 2

I never down vote or "close" or otherwise go negative, but I agree with @virmaior the only "philosophical" answers to this question will be about the structure of the question itself.

Either we are being asked to validate the simple functions of "game theory" or we being asked to validate dumping infinitely many human complexities into this function. Neither course falls, in my view, under the intermediating remit of philosophy. Here, philosophy defers to the dramatists, i.e. Miller and Mamet, in this case.

While the question of the moral status of "salesmanship" may be interesting, it must be tied to something beyond a purely utilitarian calculus if it is to be in some sense philosophically disinterested. That is, in my own view, a connection that can only be made through Marxism. Though other approaches might detour though Smith or even Machiavelli.

In any case, game theory is very significant in one logical sense, but tic-tac-toe in another. I agree that this is not yet a philosophical question. Too human (dramatists) on one side, too systems-based (machine logic) on the other. The struggle since Kant has been to mediate the two.

Answer 3

In a recent episode of EconTalk, Robert Frank (in the transcript as "Guest") gives this insight into economic-behavior problems:

I tell students who offer explanations like that to hear alarm bells going off in their head when they realize that they've just offered a cash-on-the-table explanation.

So, you might be right, but as Cort Ammon described, it could be that you are missing some other piece of context or data that someone who owns car dealerships knows. If you are right, then one could easily earn a lot of money by employing the sales incentive model you suggest.

Answer 4

In the prisoner's dilemma, there are two sides: There are the prisoners, which are in a very unfortunate situation, and there is the person who set up the dilemma, who I would assume is a thoroughly nasty character who enjoys the situation.

In your case, the individual car sales persons don't actually have a dilemma: It's clearly in every single sales person's best interest to sell as much as they can at the expense of the others. There may be some amount of back stabbing involved, but that doesn't make it a dilemma.

The other question is whether the company which set up these rules is doing something wise (that furthers their own interest). As discussed, it doesn't seem to optimise the amount of sales with the existing team, but it might force the less ruthless sales people out and increase the "quality" of their sales team that way.

So while game theory is relevant, the prisoners' dilemma isn't.