Theory or Models of Implicit Collusion

I consider the term "implicit collusion" to be an oxymoron: you can't have collusion without an agreement, and an agreement is by definition explicit. Usually the term is applied to independent actions by industry leaders who have fairly obvious common interests in an oligopolistic setting.

For example, big companies who dominate a market are reluctant to get into pricing wars with each other, since all such parties have deep pockets and are unlikely to be driven out of business; instead, the result would be a long period of decreased profits for all, and an uncertain redistribution of market share.

If one giant increases price (or sells less product for the same price -- a sneakier but increasingly common practice), because market research suggests this is possible while gaining profits, other giants likely have the same research available. If one giant raises prices or cuts product quantity, others might be more willing to take the risk, because they want bigger profits too; and if the strategy backfires their main competitor will also suffer failure since they have already committed to it.

I don't agree that it is an oxymoron.

Collusion is where there is a set of actions (or strategy) that benefit two or more independent actors as long as both of them continue to follow them. This doesn't require an explicit agreement. It only requires that both sides understand that failing two take these actions will hurt all parties involved.

In business, there are plenty of examples in pricing, growth and product choices.

In poker there are interesting examples. Explicit collusion in poker (making an agreement to act in each others interest) is cheating.

But there are a couple of plays that everyone understands and makes instinctively. The most common example is one one player goes "all in", the other players (who aren't clueless) will refrain from making any further bets. The reason for this is that it is in the interest of all other players (except for the one who is "all in") to not push other players out of the hand. This is completely legal of course... in fact this implicit collusion is considered to be proper play.

Here's a definition of "collusion" from dictionary dot com:

1. A secret agreement, especially for fraudulent or treacherous purposes; conspiracy: Some of his employees were acting in collusion to rob him.

2. In law. a secret understanding between two or more persons to gain something illegally, to defraud another of his or her rights, or to appear as adversaries though in agreement: collusion of husband and wife to obtain a divorce.

Your poker illustration is not a counterexample: independent actions based on a common understanding of the circumstances do not constitute collusion unless some prior explicit agreement had been reached.

Sure, you are correct about the definition of "collusion" as far as "explicit collusion".

But implicit collusion functions in much the same way as explicit collusion, which is why it is so named. The results are the same. Generally we consider competition "fair" as long as all the parties act in their own interests within the competitive field of play. In implicit collusion the circumstances, as understood by the colluding parties, mean that the correct strategic action is for them to act according to shared interests (instead of acting competitively).

I think my poker example is a good example of this. Let's say that play "A" is "all in" and that players "B", "C" and "D" are participating in the hand. Let's say player "B" has a good hand and the proper competitive play would be to make a bet. He won't because there is a general understanding (among all good poker players) that it is better for all of them to disadvantage player "A" rather than to compete with each other.

I contend that this is a good example. Implicit collusion involves not competing with some usual competitors to disadvantage others.

I contend that the term "explicit collusion" is redundant; and that the term "implicit collusion" was coined (and is perpetuated) by individuals with a poor command of English and a weak enough grasp of logic not to recognize contradiction, or to imagine that it is somehow negated by the persistent misapplication of language in the service of ill-conceived, unneeded, and confusing neologisms.

Since you no doubt disagree on this point also, I bid you good day.

That seems a little harsh. The term "implicit collusion" is perfectly understandable, and it is the best term I have found for an important concept in economics and game theory. That is how language works, people come up with terms to represent concepts, it makes it a lot easier than to have to describe the concept every time.

How do you feel about the word "eggplant"?