rogerd - 3:44 pm on Jul 19, 2001 (gmt 0)

US anti-trust law is funny stuff. We seem to penalize activities (like price collusion or bribing third-world generals) that other countries seem to accept as part of doing business. Sometimes this is bad (like when you lose the jet fighter order to someone who paid the general's brother a $5 million consulting fee), and other times good (when our unfettered market increases consumer buying power - look at Japan, for example - or keeps our businesses more productive).

Prices still get "fixed," of course. Like gas stations at the same intersection, steel companies all sell the same products at the same price. They are not allowed to discuss pricing with each other - that would be illegal. So how does this incredible coincidence occur?

It's simple. They publish their prices. If one company decides to raise prices, that firm lets the world know. Their competitors raise their prices by the same amount immediately; if one or more competitors announce that they will NOT increase prices, the first company cancels its increase to avoid losing business. Efficient and legal. In oligarchic industries, there is often a "price leader" company who usually sets the trend for everyone.

In the chaotic world of SEO, composed mostly of tiny independent contractors located around the globe, I think the probability of antitrust violations is nearly nil. The market is way too fragmented for any sort of collusion to be effective, and the targets are too small and fluid to be attractive to prosecutors. They like to go after firms like Archer Daniels Midland (fixing worldwide prices for some of their products). Besides, trying to get SEOs to agree on pricing would be tougher than herding cats.

Certainly, I wouldn't advise anyone to discuss setting the same prices in this (or any other) forum, or even doing so in any other manner. I think referring to (or even reproducing) any public pricing is probably fine, but I'll defer to Brett and his legal advisors as far as the limits he sets for posting here.