What Is A Japanese Auction? – And An Ethical Issue

What is a Japanese auction?  In the classic “retail” design, the initial price is displayed, and all the bidders who are interested in buying at that price “enter the arena” (expressing their interest in buying). The price is then increased either continuously or in increments, and gradually the bidders drop out as the price rises. The moment only one bidder is left, then that is the price at which they buy.

The Japanese Reverse auction is the business equivalent for procurement purposes. The price starts high and decreases, and the different bidders (sellers) gradually drop out as the price decreases. When the second to last bidder drops out and only one is left, then they are the winners and are awarded the contract. (Or it could be more than one left, if the intent is to appoint multiple suppliers).

However, a dispute has broken out in the procurement software world. We even hear stories of battles over Wikipedia entries! The question is: do the bidders know when they are the last one standing, and does the auction stop automatically at that point?

In the classic design, the answer is “yes” – the original Japanese auction would have taken place in a physical arena, with bidders able to see who else was bidding and when they were the only bidder left. But in cyberspace, that is not the case. So, the auction system can let the suppliers know how many participants remain and can be set so it stops the process when only one is left, who is declared the winner. Or…. not!

It is feasible to allow the last bidder to keep involved in the process as the price drops further. So, everyone else is “out” at £100 say, but one bidder continues, and the price keeps dropping until that final supplier drops out at £90, let's say. At that point they are told they are the winner. But is this unethical? The supplier “won” the auction at £100. Have they been conned into carrying on reducing their price down to £90?

Well, we can see both sides of the argument. A cynical old procurement manager like me might suggest that if a supplier is happy at £90, then why shouldn’t I take advantage of that? If they aren’t making any money at that price, why did they offer it? The mechanism is just establishing the minimum price at which they will do business with me, so what’s wrong with that?

On the other hand, not letting them know they had “won” at £100 might be seen as deceitful, and it is certainly not transparent for the supply side. At a minimum I’d suggest that it must be made very clear to suppliers before they enter into such a process that this is how it will work – they can then decide if they want to participate and how they want to implement their auction strategy.

The argument in the industry is also that this should not be called a Japanese auction and needs a new term. Perhaps a “Trumpian auction”, in honour of the President’s ruthless (allegedly) style of doing business?

The ethics of behaviour in negotiation and bidding and tendering processes is an interesting question generally. Perhaps we’ll come back to that another day, but let’s leave with a question -  have you ever lied to a supplier to get a better deal?

Voices (3)

  1. Nick Drewe:

    Definitely interesting! It does seem that generally the traditional auction design has certain nuances when migrated to the digital environment. For example, majority of private sector eAuctions are run on a ‘buyers choice’ whereby the buyer will decide who is awarded a contract and not default it to the winning bidder. This is a massive variation from your traditional auctions at Sotherby’s, Christies, even eBay, and is a bigger debate than a Japanese Auction feature.

    As a supplier, we will compete against ourselves for almost all RFPs & RFQs. We take a subjective call on how competitive the bid is, we decide how important the client is and what future potential there is with the client, we decide where overheads can be minimised by collaborating with the client, we might be given budget constraints to work towards….+ many other factors. A Japanese Auction should be no different in that regard.

    Given how many private sector decisions are based on ‘buyers choice’ it is in the suppliers’ interest that they are able put their best possible bid forward in the Japanese Auction, just as they would in a sealed tender suggested by Secret Squirrel. I’d be livid as a supplier if the Japanese Auction ended on me even though I had room to move, to be later told I was not successful with the opportunity – I had more to offer that was not considered!

    Here are some actual examples of where we’ve seen Japanese Auctions work better for letting all suppliers put their best bid forward:

    1) Discounts from Rate Cards: Rate cards may have several thousand goods or services items within them. eAuctioning each individual item is challenging for suppliers as a) forecast volumes are often unpredictable/unknown and b) there are a lot of items and a supplier may only cater for a certain subset of them. So a Japanese Auction works really well where the supplier uses their existing pricing as the start point and will then be asked to commit to an incrementally larger % discount during the auction. In this situation it makes no sense to stop the auction with the last bidder, as each bidder is using different start points (existing prices) from which to apply the % discount. So, on it goes until everyone has agreed to their best discount for the negotiated Rate Card. Also note that these may not even be the same rate cards and could involve suppliers across dozens of supplier categories, as it’s just a % discount that is being sought.

    2) Negotiating rebates: Similar to the rate cards, this could involve suppliers across many different categories. It could even be the entire supply chain, as one client considered (500+ suppliers at once). The Japanese Auction can work really well for a simultaneous negotiation, again using a % rebate agreement. And again, it must allow all suppliers to present their best offer.

    Hope that provides a bit of substance on why it can be useful to all parties to let the Japanese Auction complete once everyone has been able to reach their BAFO, and that really it’s just a micro feature switch to enable/disable single supplier bidding in a Japanese Auction format rather than a whole new definition!

  2. Jason Busch:

    It’s a definitional issue … anyone calling it a Japanese Auction should adhere to the standard definition, lest a supplier be mislead. The model that allows suppliers to compete against themselves is not a Japanese auction by definition!

    1. Secret Squirrel:

      May I suggest that this is no different than a sealed bid tender? You decide on the price you’re prepared to bid, you just do so in a manner where bidding down on yourself is visible to the buyer.

      Also, in a perfect information world all auctions have the same expected revenue (or cost). How you get there isn’t a function of the auction method, it’s just a function of there being a competition. So a second price auction will have the same expected cost as a lowest cost auction. Similarly, a Japanese auction has the same expected outcome as an English auction. What happens is that bidders change their bidding behaviour so the overall outcome ends up roughly the same (on average).

Discuss this:

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.