A similar thing happened to me as a seller. I saw that one of my old textbooks was selling for a nice price, so I listed it along with two other used copies. I priced it $1 cheaper than the lowest price offered, but within an hour both sellers had changed their prices to $.01 and $.02 cheaper than mine. I reduced it two times more by $1, and each time they beat my price by a cent or two. So what I did was reduce my price by a few dollars every hour for one day until everybody was priced under $5. Then I bought their books and changed my price back.
The opposite scenario (sort of), from Michael Lewis' Liar's Poker:
One day earlier in his career Dall was in the market to buy (borrow) 50 million dollars. He checked around and found the money market was 4 per cent-4.25 per cent, which meant he could buy (borrow) at 4.25 per cent or sell (lend) at 4 per cent. When he actually tried to buy 50 million dollars at 4.25 per cent, however, the market moved to 4.25 per cent-4.5 per cent. The sellers were scared off by a large buyer. Dall bid 4.5. The market moved again, to 4.5p per cent-4.75 per cent. He raised his bid several more times with the same result, then went to Bill Simon’s office to tell him he couldn’t buy money. All the sellers were running like chickens.
“Then you be the seller,” said Simon.
So Dall became the seller, although he actually needed to buy. He sold 50 million dollars at 5.5 per cent. He sold another 50 million dollars at 5.5 per cent. Then, as Simon had guessed, the market collapsed. Everyone wanted to sell. There were no buyers. “Buy them back now,” said Simon when the market reached 4 per cent. So Dall not only got his 50 million dollars at 4 per cent but took a profit on the money he had sold at higher rates. That was how a Salomon bond trader thought: He forgot whatever it was that he wanted to do for a minute and put his finger on the pulse of the market. If the market felt fidgety, if people were scared or desperate, he herded them like sheep into a corner, then made them pay for their uncertainty. He sat on the market until it puked gold coins. Then he worried about what he wanted to do.
I imagine there is a lot of money to be made on Amazon using this Salomon bond trader technique of simply putting a finger on the pulse of the market.
In calculus terms, it would be like taking a derivative of the Amazon marketplace and operating on different rules than most or all buyers and sellers on the marketplace.
The trick is minimizing your risk and making sure to adhere to Amazon's terms. [Does the scenario in the grandparent comment go against Amazon's terms I wonder?]
If you play World of Warcraft this is a solid Auction House technique. I find it humorous to see it applied to the 'real' world (not that WoW economics aren't just as 'real').
The ability to push the price down gives informed actors some pricing control on the market, and the subtext is that it allows for folks to push around pricing to understand the elasticity for any product. Now if you can get them to buy a 'put' of your product :-)
I'm not a lawyer, but I've worked in the financial sector and undergone FSA mandated training courses on what you can and can't legally do with equity pricing. What the OP was suggesting is an illegal practice in most equity markets.
The reason it's "liar's poker," not "outright fraud," is that if someone calls your bluff, you lose that hand. The successful traders were just good or lucky enough to be able to bluff the entire market.
> And (from what I understand) this is just the kind of thing that High Frequency Trading algorithms are doing.
In stocks, there's almost always someone willing to buy and someone else willing to sell at any given price, so how are you going to drive the price down without selling stock?
That is the most fascinating myth I continue to hear about the market. As a non-seat owner, you see three prices: The price of the last trade OVER 15 minutes ago, where a buyer and a seller agreed to buy. The highest price someone offered to buy the stock 15 minutes ago, and the lowest price someone offered to sell the stock 15 minutes ago. When you actually buy or sell a stock, you often do not buy from or sell to a long term owner. Instead, the transaction happens with a "market maker" an entity that is a seat owner, knows the current offers right now as opposed to 15 minutes ago, and trades with you. There is almost NEVER a "meeting of the minds" on price between a buyer and a seller. In fact, there does not have to BE a buyer if a market maker decides the value of the stock makes it worth acquiring or selling at your offer price. (If you check "Market" for your offer price, I suggest you read almost anyone who writes about trading, but plug your ears because they start shouting "no" really loudly). So, to finish (sorry about the length of this), a huge or in some cases a simply mis-priced offer to buy or sell may unnerve the market makers. They may hesitate. The lack of quick sales at a given price to the market makers causes repricing by the sellers, which causes the price of the stock to fall. Try pricing a limit bid between the buy and sell price on your electronic brokerage. You'll see. Sometime it will sell right away, and sometimes it won't. If you want to sell right away, you'll drop the price. And, if you happen to find a stock in "free fall", you'll never sell, because the price you see is 15 minutes old.
Please update your facts. Almost every online broker offers real time pricing for US equities and some even provide the depth of book. What you say above (15minute delayed quotes) was true several years ago but things have changed.
not quite. What OP described would not work in automated finance, where once the book approaches $5 an army of computers would try to buy it before him.
This is possible because the algorithms running are still immature and the creators either didn't setup any sanity limits because they are inexperienced or maybe lazy.
I wouldn't expect this to last very long, but you could potentially take advantage of it while it does by simply identifying seller's who lack limits and posting products within their categories at super low prices then buying them out and re-listing.
Eventually all of these scripts will have guards in them for upper and lower and probably limits on the percentage change over time.
I love that. It's a great lesson to programmers that we should think about the extreme cases and have set lower and upper bounds on our algorithms. Sanity checks should be in your code for most applications and inputs.
Amazon should put an upper bound on the maximum price a product can be. Can credit card companies even process a transaction in the tens of millions of dollars?
You could turn this into a book business. List books for sale that you don't own and this other publisher does own. Reduce the price over time until it is super cheap. Buy while it is low and resell it without the competition.
It is left as an exercise for the reader to show how to determine whether the other publisher actually does have a copy, or is indulging in similar behaviour.
This reminds me of a similarly unethical behaviour I've encountered recently: I tried to buy a certain rare electronics component, and found it listed by a bunch of people. There were lots of bait-and-switchers in the Google results, who only had the part listed for the googlebot to pick up, but as soon as I actually tried to order the part, their site told me they didn't actually have it and suggested I buy something else from them.
It gets worse: When I finally did find a seller who allowed me to order the part, they accepted my order -- and then came back later saying that, oh no, actually they didn't have it either, but would I like to convert my order into an order for something else? (Fuck off.)
My wife has experimented with that, sort of. She wouldn't not try to force the market down, but she'd occasionally find arbitrage situations on Amazon and exploit them. Buy a book placed at far too low of a price and then simply re-list it at a correct price.
The thing is that while these automatic pricing scripts exist, they aren't in enough widespread use to do what you're suggesting.
Oh, but in naked puts the seller of the underlying assets (i.e. the buyer of the put) calls the shots. If they don't want to sell, they don't have to sell. And the writer of the put only ever has to put up at most the strike price of the put.
That's unethical business behaviour - it a microcosm of monopolies: buy up all the supply, then sell at a higher price than the previously going rate. It's common in business, obviously, but it's still unethical.
Not to mention that in order to make money over normal practices, you have to a) find books that are being sold at less than wholesale price (~60%ish RRP), and b) find a book that no-one wants to buy at your competitor's super-cheap rate that they will buy at your more expensive one - you're going to have to be in for the long haul to do this.
I don't think this strategy is unethical; it is really risky... probably so risky that it's naive for the long term.
It's not unethical because the buyer in this case can't really buy up all of the supply and therefore is just looking for arbitrage opportunities or opportunities to exploit weaknesses in a competitors business model.
The opportunities, over time, are very limited though:
1) You can't actually buy up all of the supply. If you were able to buy all of the listed supply and then reset the price you'd soon attract other sellers who had previously unlisted supply.
2) If there was real demand for an extremely high priced book I'd guess the author or publisher would be releasing a new version to meet the demand.
3) Although some printing of books are considered to be collectible, that market is limited. Generally the value of the book is the information it provides, which means that as the price rises the (perhaps less informative) substitutes become more appealing to buyers.
4) If you exploit a competitor's pricing algorithm enough to make them notice, they'll just change their algorithm. In this case, perhaps they don't sell you the books or they don't respond to your prices.
You might be able to make a few bucks on this strategy, maybe a few people could even eke out a living, but it would be risky... they'd earn it.
Not really that risky. If it was a naked call like in markets, then it would be really risky, but in this case you risk very little since there is no contractual obligation to send the actual book if you can't get it. You can reimburse and call it a day.
The calculations would have to consider the probability of not getting the book under the price of sale, the probability of not getting it over price of sale (investing the difference in avoiding a bad rating), and the chance that you won't receive a bad rating just for reimbursing it under whatever excuse. You can also ponder taking a bad rate eventually.
In other markets you'd have to basically get the book at any cost, which would make it really risky.
And in fact, in used book "arbitrage", the seller always has a very, very good excuse, which is that s/he "sold the book already" to someone in s/her imaginary brick-and-mortar store.
I was mostly joking but if you were really going to do this you'd have to develop strategies to reduce the chances of this happening and mitigate it when it does.
Maybe if you have a slightly bad reputation on Amazon that will keep people from buying your book. Maybe you can purchase the book locally and send it to the buyer in a pinch. Maybe you can purchase the book from the actual seller and just use the buyer's shipping address. Etc.
Why go to that effort? Did you really save any money, or more importantly time/effort? Paying for the shipping on the other books alone would seem to negate any higher profit margin that resulted in not having to compete with their prices.
This explains a lot. I wrote a small book back in the 90s that had few sales and was unlamented when it fell out of print. A couple of years ago saw it as the subject of an amazon.com sidebar ad and was astonished to find that it was listed at $100 or thereabouts, and couldn't imagine how or why it might have become collectible. The idea that this was the result of competing pricing algorithms makes a great deal more sense to me.
Now, if you'll excuse me I'm off to weep over my broken dreams of belated celebrity.
My dad had a similar experience. His very obscure, self-published, out-of-print book was showing up for something like $1000, while at the same time he was mailing out free copies to people who emailed (plus shipping), just to make more room in the garage.
I work at a mid sized Amazon 3rd party seller. We reprice automatically. With thousands of SKU's there's no other way to be competitive. There are many layers to consider though, you definitely can't let your whole catalog auto-reprice. Usually sellers just focus on their top X% SKU's and let the rest auto-calculate.
I wonder if we have a price ceiling setup...?
Edit: Yes we have a floor and ceiling. I have no idea what I was thinking.
I have code that does ecommerce price point optimization using machine learning. In an ideal setting, it combines a variety of signals, including prediction using previous sales data. It's also able to "smooth" over different products, so that you can make good price point choices even for products that have very little sales history.
Probably some sort of regression for the prediction part, and perhaps linear programming for the price optimization, since there are probably some nontrivial (but possibly linear?) constraints on pricing.
What would be even more useful is to have the browsing customer's data passed in as a signal as well, but I'm guessing Amazon doesn't send that out to third parties. If 3rd parties could also customize their prices based on the potential of someone to buy (and perhaps offer bundled items based on their own algorithms) we'd probably see even greater efficiencies than we've already seen.
Could be this: If two sellers have the book, and are setting the prices algorithmically, and then one of them sells it offline (like in a used bookshop) and removes it from sale, how does the other seller change their price to be more realistic?
oh man. reminds me of the days when I sold books on amazon and half.com. I wrote a script that took the 'nickel less than the other guy' approach.
These things are /wonderful/ when it comes to making the market more efficient. Really, though, the shipping costs eat up most of the efficiency. Amazon needs an easy way to say "I want these 10 books, used. Find me the lowest price (including shipping)" - the idea is that the more books you could buy from one seller, the less shipping friction would be involved, but amazon isn't really set up that way, which makes it much less efficient for the low end used books.
Everyone always mentions that buying more items from a single Amazon store would make shipping cheaper? Is that ever true? Of all the online stores I've made purchases from, I believe monoprice is the only one that appears to base shipping on tiers based on item size/mass. All the rest simply charge a set rate for each item, and it doesn't matter how you divide your shipments up into different orders.
well, the post office would take less of a cut. shipping one one-pound box is usually a lot cheaper than shipping two half-pound boxes. In theory, amazon could either give that surplus to the seller, to the buyer, or they could split it between the two. (I guess they could also keep it for themselves, but that might have negative PR consequences.)
Amazon gives you the option to group things to reduce shipping. And if you do this you do get a discount and you will revive a single box with several items in it. If you are ordering small and cheap items like USB cables the discount can be significant.
Another very common system is to base the shipping cost on the total order price. Not very accurate, really, but mainly used because it's simpler for people to handle when filling out the order form that comes with a paper catalog.
There seems to be a reasonable amount of literature on algorithms for this kind of combinatorial optimisation, including reasonably fast but well-performing heuristics like the above. Maybe amazon should pull their finger out.
Very funny and enlightening analysis. I've seen the fringes of algorithmic pricing a few times in other categories -- especially out-of-print DVDs or VHS tapes.
Example: don't ask me why I enjoy the movie My Dinner With Andre, but for whatever reason, I do. A few years back, I wanted to buy a copy of the DVD and checked Amazon. It turned out that the DVD was long out of print, and new copies were going for $400 apiece. I figured this price was high, but nevertheless, it was nothing that couldn't be explained by actual rarity and supply/demand metrics. Rare DVDs have been known to climb into the hundreds of dollars, especially if new and unopened. But I came back a few days later, and the price was $1932.78 (or something unusual to that effect). The next day, $3500 and change. Which struck me as odd, to say the least. Was some nefarious Goldman trader attempting to fix the market for Wallace Shawn's back catalog?
Needless to say, I didn't love the movie quite that much. So I passed. These days, Criterion has released a new DVD version of the film, and accordingly, everything's dropped back down to about $30 per copy -- including the price of the original, OOP version. I feel sorry for anyone who actually might have taken the plunge at $400, which is not out of the realm of possibility. That's a lot of money for a film about two guys having dinner.
Alas, while it seems that information technology-wise we delve more and more into a cyberpunk world, health-wise and body modification-wise we are still not there. Where are my grafted muscles and commodity eye-balls?
Automatic pricing is super common these days, even on more expensive things than books. I'm actually working on a startup called shopobot.com that wants to help people use the volatility to their advantage. We see rapid $50-100 swings on things like SLRs, so it's actually pretty significant.
Ironically our site is down right now because we're based on Amazon's web services. Karma? :)
That actually sounds pretty awesome. Didn't some of those CSE's like MySimon used to have a graph of pricing over time as well? Any way I can get an invite code for Shopobot?
I have been an on line book dealer for the past 10 years. Originally the rules about having actual possession of the goods offered for sale were quite strict.
Then came the megalisters - agencies with software that listed all books in print and a contract with publisher's warehouses/library supply services/factors to arrange delivery of in print books to customers.
Then came the "phantom listers" - agencies with software that spiders through the listings of legitimate on line book dealers looking for titles with few (or no) copies listed in Amazon. They then list them at inflated prices. Some of these have more than one alias on Amazon.
I think of them as the Piranhas of the Amazon in the ay that they consume the smaller fish.
If and when they receive an order then then try to purchase a copy from one of the "real" owners of the book.[Some are cheeky enough to request as well a"trade discount" on the dealer's price] Small businesses listing on Amazon and abebooks.com must keep their "fulfillment"rating high, so they can ill afford to refuse to supply such parasites.In fact when I requested once that I NOT be obliged to support what I think of as an unethical selling and pricing model I was told
1] that the buyer was a valued long time customer
and that
2] I could approach the agency in question and - at THEIR discretion - request that they stop ordering books from me.
Since AZ and ABE take out some % of the list price for the the books sold on their site (as well as monthly listing fees, closing fees and sometimes a portion of the shipping charge ) they can and do make a much greater profit from the high priced phantom listers than from the legitimate reasonably priced offerings of small book dealers, consequently they are not very interested in aggressively policing the situation.
Now there is "Monsoon" and other software to automatically adjust prices on line. In practice most often works to REDUCE the price below the lowest price already listed - a rush to the bottom where books get listed for mere cents.
One tactic I recommend is to search for books and to see the widest range of options is to use "addall.com" It comes in 2 flavours "New' and "Used"and searches around 30 book listing services You can compare prices (ascending or descending) and also see the kind of dealers who sell the books.
Many listings have boilerplate descriptions "we ship fast" "books may have..." etc. which indicate that no human may have examined the object being listed.Other listings have descriptions of content and condition that clearly demonstrate that it is a"real"book from a "real"dealer.
When in doubt look for THAT dealers OWN website to ask questions and get personal service.
Hmmm. I wonder if this would work in reverse? Suppose I want to buy a book that has several new copies for sale on the Amazon Marketplace for say $60. If I post a new copy for sale for $10, perhaps one of these algorithms would kick in to reduce the price?
Its all innocent when it is Amazon. But I am sure there must be similar phenomena taking place in the financial markets, but the algorithms that determine pricing there are more complicated than multiplying by a simple factor.
bordeebook was effectively increasing their price by a factor of 1.2684/day, and as of Apr 8 priced the new book at $2,198,177.95. I'll guess that the used price ($35.54) implies that the initial pricing of the new book was about $50.
Log base 1.2684 of the ratio of the Apr 8 price to the estimated original price = log($2198177.95/$50)/log(1.2684) = 45 [days]. Ergo looks like the two algorithms started competing around Tuesday, Feb 22.
I'm sure my original price is off, but that actually won't make much difference. If it was $40, then the initial date was Feb 21; if $60, then Feb 23.
[EDIT] By the way, if the price was originally 99c you would still see the $2.2M price after just another two weeks (61 days total). Ah, exponents.
The original price candidate, derived from the logarithm, that is second closest to integer cents is 62.91998, with 162.85966 coming in 6th and the other positions filled by implausibly large numbers.
Neither number has integer cents, since the actual series rounded or truncated values between iterations. For that reason, a simple logarithm won't tell us what the original price was - you'd need a simulation of the rounding.
Good idea! Here it is in C: http://pastie.org/1824968
(You have to manually add the 'f' after the constants if you want to do the whole thing in floating-point, as well as uncommenting the correct Dollars_t definition.)
For the range $20.00 to $200.00, no exact matches to an integer original price (in cents) for the algorithms ceiling, floor, or round, in float or double. This was on a 32-bit Xeon server (EC2).
> My preferred explanation for bordeebook’s pricing is that they do not actually possess the book.
There's only one problem with this theory. The Amazon TOS requires sellers to have inventory before they can sell inventory. And they will ban you for life for violating that.
Yes and no. It's true this a policy violation, but I can tell you unequivocally that a huge percentage of used sellers on Amazon are actually arbitraging their competitors inventory. As a midlevel sized buyer and seller on Amazon, I've always assumed that these sellers had some kind of secret upper-tier arrangement with Amazon? Because as buyers, we see high volume sellers do things on a daily basis that we would never do as sellers for mortal fear of our account getting banned. Yet somehow these megasellers seem able to muddle on year after year?
A lot of them are not necessarily even primarily e-commerce sellers, but are semi-camouflaged divisions of larger traditional publishing entities - like how the seller oneworldbooks is actually the textbook wholesaling giant MBS Books. These sellers are companies so large they are akin to major new-book publishers, so I assume their relationship with Amazon as a company is cozier than ours is as "random small bookstore with Amazon account".
There is an upper-tier "club" on Amazon, you can see it alluded to indirectly in various SC help files. Maybe these guys get away with more than we do.
Ha ha ha ... I've encountered lots of annoying sellers who do this -- I make an order and they cancel it last minute because they obviously don't have the book/CD. Yet I keep on seeing them appear.
Pricing something high can actually increase sales. There is an example in Cialdini's book "Influence", near the beginning of chapter 1. A jewelry store owner was trying to get rid of some items, instead of marking the price down, she marked them up (by a factor of 2), and they were rapidly gone.
It is a nice book to read, and will probably make you feel uneasy when you realize the abundance of manipulation tactics around us.
[Discalimer: I've never looked into this before and I have absolutely no idea how the Amazon reseller market works, so this might be impossible or prohibited.]
What's stopping somebody from relisting every product that already exists in the reseller market, but raising the price a few bucks? Even if people are more likely to buy the lesser priced product, the seller has to lose since they don't actually have any products on hand or skin in the game. Better yet, they could relist the products for less than the competitors and jack up the shipping prices to skim a few bucks profit off the top of each sale (this would work if somebody sorted by item price only, instead of price + shipping).
The risk is that you're on the hook for the book. You can certainly drop ship, and it's a common practice, but you have no quality control. For example, if you're drop shipping a book in "Good" condition and it gets to the buyer in "Fair" condition and they complain, you're out the money and it's your feedback on the line.
From experience, overpricing works when you have better feedback than your competitors.
My first question: what's the penalty for listing something that you can't deliver? As in, if someone buys 'your' copy, adn someone else buys the only other copy before you get it, what do you do and how does Amazon react when you do that?
Furthermore, though, how much more do you have to charge to cover your own time, shipping complexities, and just the general mess of dealing with two other entities for every sale you make? All of that adds up to a pain in the arse to make a couple bucks, in my estimation.
I've been a seller on amazon for the last 6 months. Your order to cancel ratio needs to be below 7%. If it's above this, your account will be restricted (you won't get paid during this time). It happened to me...but it was lifted after about a month.
Depending on the category, many things on Amazon sell so fast, it would be very difficult to do this.
>Furthermore, though, how much more do you have to charge to cover your own time, shipping complexities, and just the general mess of dealing with two other entities for every sale you make? All of that adds up to a pain in the arse to make a couple bucks, in my estimation.
If you can operate at large enough scale, thousands or tens of thousands of books a month, then I think there's an opportunity to be had, even if you're pocketing a dollar or two from each sale.
I've seen lots of books on Amazon priced at $0.01 or $0.02 before. Maybe a result of similar processes? (I always assumed they were some kind of scam, since why the hell would someone sell a book so cheap?)
Those sellers are counting on making money by shipping for less than the amount by which Amazon reimburses them. ($3.99 for a recent purchase of mine, while the postage was about $2.00.)
They are. I sold a chunk of my library, a couple hundred books, on Amazon, and saw sellers who would update their prices to be a cent or two below mine any time they didn't have the lowest price. The other commenter is correct, that there's still profit to be made in shipping charges.
Sometimes you can use this for trade ins to Amazon directly. Last week I traded in a book that had several prices for $100.
Amazon took it for $27. On other sites you can get it for $24.
It won't, because although computers have a lot of control over the market, people also look at it pretty regularly. The reason why nobody noticed a $23,000,000 book on Amazon is because out-of-print biology books are not very liquid. If that were Microsoft stock, someone would have manually made a bet on it, and then the price would have corrected after a while.
No. If someone were asking $23m for a share of MSFT it would just be way outside the market and would just sit there. Plenty of people leave way outside the market bids and asks just in case someone else makes an accidental huge market order.
True, but at the same time, we're talking about billions of dollars of real money. In clear self interest, all those institutions will have built obvious checks in their system. Losing several billions of dollars in seconds is something they want to avoid.
Indeed, the fact that the flash crash was the definition of something with a vast number of transactions, huge price moves, and happened in the blink of an eye but still was relatively minor and almost inconsequential shows the reliance of the stock market.
Someone should try selling a book that there are limited copies of (ie one that can't be procured easily) to see what happens if someone buys from a seller who doesn't have it.
My wife used to sell used books on Amazon part-time a few years ago, mostly as a hobby to feed her own library. She employed some software and pricing algorithms, similar to the ones mentioned in the post.
Often, she'd go to library or estate sales and just scan hundreds of books. If they met her pricing targets, she'd buy them. But one time a book she scanned had an incorrect barcode number (it wasn't clear if Amazon had the incorrect barcode or what), but so did several other sellers of the same book.
When she posted the book for sale, it came up with a different title, but she hadn't noticed while she was posting inventory. Initially my wife assumed she'd lost the inventory when someone tried to buy from her, but it wasn't until she tried to drop ship from another seller that they realized what was going on.
The sellers all had to cancel their orders with each other on down to the buyer, who took it in stride.
But, with Amazon, you can still rate sellers who cancel your order, so there's at least that risk...
The nice thing about selling books is that you aren't bound to deliver the book - you can always just cancel the contract and refund the money. It's unlike short selling stocks in that respect; you aren't vulnerable to a short squeeze (http://en.wikipedia.org/wiki/Short_selling#Risk).
The NYT piece on Porsche & Volkswagen (and many other market corners) is actually a really useful piece of financial literature. It nicely outlines some of the outlandish and clever ways that people can use common market forces to create situations for profit.
I always try to think in a similar way when it comes to startups. Where can you harness some market force to create a favorable position for yourself?
The seller will drop ship it from another seller or they cancel the order. Not a lot of options. As long as they don't do it too often it won't affect their feedback ratings noticeably.
