A few weeks ago, a doctor from my laboratory went to Amazon to buy a copy of Peter Lawrence’s
The Making of a Fly — a classic work on developmental biology that we — and other biologists working with fruit flies — regularly check. The book, published in 1992, is no longer reprinted. But on Amazon there were 17 copies on sale: 15 used from $ 35.54 and two new ones, from $ 1,730,045.91 (+ $ 3.99 for delivery).

I sent a screenshot to the author of the book - who was very surprised and intrigued. But I think even he would not insist that the book is so expensive.
At first I thought it was someone’s joke - some kind of former student who has nowhere to go. But the sale was just
two copies, each for more than a million dollars. And the two sellers looked not only quite honest, but also fairly well-known (more than 8,000 and 125,000 ratings for the last year, respectively). Prices looked random - it looks like they were set by a computer. But how do they achieve these values?
To my surprise, when I reloaded the page the next day, prices rose even more! Each book cost almost $ 2.8 million, the price difference was less than $ 5000. Now I became interested in the situation, and began to closely monitor the page. By the end of the day, a more expensive copy rose in price even more. Now it was worth $ 3,536,675.57. Finally, the overall picture began to emerge.
')
On the day when we discovered multi-million dollar prices, a copy from
bordeebook was 1.270589 times more expensive than a copy from
profnath . And now a copy of bordeebook was more expensive than a copy of profnath In 1.270589. Obviously, at least one of the participants put up a price automatically in response to a price change from another. I continued to observe and soon I understood the whole process.
Once a day profnath set the price 0.9983 times the price of the bordeebook. Prices stayed in this position for several hours, until the bordeebook “noticed” the change in the price of the profnath and did not raise its price by 1.270589 times the maximum price of the profnath. In this order, the events continued in the same order throughout the next week.

But there were still two questions. Why did they do it and how long will it last until someone notices? While I was surprised to see the daily price increase, I learned that Amazon sellers are increasingly using price setting algorithms (Amazon itself does the same on a much larger scale), with several companies offering services / algorithms to sellers to determine the optimal price. Both profnath and bordeebook explicitly used automatic pricing — while using algorithms that do not have a built-in sanity check of the set price. But these vendors clearly used different strategies.
Behavior profnath is easy to understand. They most likely have a new copy of the book, and they want to be sure that their price is the lowest, but not much different from the next ($ 9.98 versus $ 10.00). Why did bordeebok set the price higher? Since the system shows the prices of all sellers, it seemed that they were guaranteed no sales. But perhaps this is not entirely true - they have a huge amount of positive feedback, much more than the rest of the sellers. And some buyers are probably willing to pay a few extra dollars for the confidence that the purchase will indeed be successful. Anyway, this is a rather strange and risky strategy to rely on - most people most likely do not act like this - and the book will collect dust on the shelf all this time. Naturally, except in the case when you really don't have a book ...
My explanation for the behavior of a bordeebook is that they really don't have a book. Instead, they noticed that someone else had put the same volume up for sale, and put it out too - hoping for more feedback to attract customers. Naturally, if someone orders a book, they will have to get it - so they had to set their price more - for example, 1.27059 times more - the price at which they would buy this book from another seller.
Most of all in this situation the endless amount of possible chaos and problems surprises. It is unlikely that we stumbled upon a single example of a constantly rising price spiral - and all it took was two sellers, changing their price according to the price of the other in proportion, ultimately leading to an overall price increase. And although it may be harder to calculate, you can think of other strange scenarios that can occur if the number of participating vendors becomes more than two. When I realized what was happening, I and some people with whom I spoke began to invent ways to use our ability to predict the seller’s price setting to five decimal places - especially when the price was set without checks.
However, soon someone noticed the problem. The price reached a peak on April 18, but already on April 19, the price of profnath dropped to $ 106.23 and the borderbook soon set the predictable price to $ 106.23 * 1.27059 = $ 134.97. But Peter Lawrence can now safely brag about the fact that one of the largest and most respected companies on Earth estimated his book at $ 23,698,655.93 (plus $ 3.99 for shipping).
