The Meaning of Collusion

Skeptics of the market love conspiracies; it's an easy way to side step the observed and provable to make outrageous claims. To no surprise of my own I came across a few conspiracy-laced theories Friday night, the most notable of which was the "implict collusion between car manufacturers."

What? I said. We see intense compeition between such firms; how can one claim they are secretly colluding?

Well, the AU student in question used to work in the industry and noticed the profit margins for car parts and vechiles was "identical." He concluded there's an acceptable level of profit and they collude to keep that profit steady.

I bypassed the obvious point that if they are setting the price they collectively could increase it easly suspecting he would respond with the standard "keeping up apppearances" defense. Instead, I pointed out that uniform prices are not evidence of collusion (though I don't doubt such prices are more or less uniform). An enforcement mechanism designed to hinder defection is evidence. What is the enforcement mechanism for car makers?

His answer was vague, claiming firms would be kept out of "special deals" if they defected. Backroom deals. Secret agreements. Implict collusion. It's not that I don't think conspiracies can't exist, it's just I'll only consider them a possibility if there lacks a better explanation.

To explain uniform prices, I noted the role of risk. Just as oil companies experience high profits and high losses (but unremarkable profits when the quarters are averaged), car companies probably demand higher mark ups because they deal in durable (and thus riskier) goods. At least one AU listener acknowledged the role of risk in determining profits but my fellow conversationalist did not seem to agree.

NOTE: My AU counterpart also cited the perfect competition model as evidence of collusion; in "perfect competition" there are no profits (which is not true; there are no windfall profts). Thus, the industry is not perfectly competitive. I told him it's distrubing when PhD econ students think a model completely describes reality. PC is useful to explain some phenomena but as Hayek noted, the model lacks all elements of actual competitiveness. He mumbled something and let the issue drop there.