Mike Eisen is a professor in my department, and he had a great catch the other day. He wanted to procure an out-of-print book on flies, and noticed that while there were many used copies available via Amazon, there were two vendors selling new copies for absurdly high prices. Read his whole post, as it is a great example of observation and deductive scientific reasoning. He noticed that one vendor would price their book at 98% the cost of the competitor, which is a strategy we can all understand. The curious part is that the other vendor would respond to the pricing by increasing theirs to 1.27x that of the first. Because 0.98*1.27 > 1, the prices were converging towards infinity.
But why would the second vendor do this? Eisen offers two hypotheses. One, the second vendor has a lot more reviews, and thus might be charging more because consumers might be willing to pay more to obtain a book through a vendor they consider more reliable. He finds a second possibility more likely: that the second vendor doesn’t actually possess the book. If someone were actually to purchase it from them, they would need to get it from another source, such as the first one, and thus the need to charge 127% their price, with that extra 27% being the profit they’d make. I can think of a third hypothesis. It’s possible that both vendors are owned by the same entity. If you only see one new book price, it can be tough to know if you’re getting a good deal or not, and you might be tempted to get a used copy instead. However, if a dummy second higher price were also listed (i.e. there is still only one new copy of the book in existence), you might now think that the first price is a deal after all. Like Stendhal says in his previous post, our concept or “saving” money now basically means “spending marginally less for a discretionary purchase.”
Also, this story underscores the weird possibilities that can arise from using automated algorithm-based pricing on large websites like Amazon. This is surely not the only time this has happened. On Reddit, someone saw a CD selling for $2.9 billion, and he actually bought it! Amazon responded by canceling the order, and following up with emails and phone calls to verify the matter. While these two examples might seem strange or even funny, it’s more alarming to think that similar automated pricing is probably largely contributing to our entire stock market and thus to the world economy.