A proliferation of penny auctions

Last year I thought about blogging about Swoopo, the "entertainment shopping" site that is run as an "all pay" auction for consumer goods, in which all bidders must pay to bid, but only the winning bidder receives the object. But soon enough there were excellent posts on the subject by others. I particularly like Ian Ayers at Freakonomics, who explained how Swoopo is similar to the "escalation auction" proposed years ago by Martin Shubik, which has become a staple demonstration in game theory classes, and Tyler Cowen at MR, who observes that Swoopo auctions can make a lot of money for the owners of the site, while most of the bidders lose. He writes "In short, swoopo is about as close to pure, distilled evil in a business plan as I've ever seen. " (emphasis in original). And here's the Wikipedia entry.

Swoopo bidders are a bit of a puzzle of the behavioral economics kind: are they like buyers of lottery tickets, who know that they will likely lose but find entertainment value by purchasing the right to dream (see this paper by Emily Oster)? Or are they making mistakes? And if the latter, will demand for this kind of auction dry up? Or will new suckers keep appearing?

But there are other, market level questions we can ask, and I got the beginning of an answer when I did a google search on "swoopo", or another search on "penny auction" . You'll find two things if you click on those searches: there are now a lot of similar auction sites, and there are also plenty of people who are eager to sell you software to set up your own "penny auction," as these sites have come to be known.

(BidRodeo's icon is a man on a bucking bronco, over the motto "Hold on the longest and win!")

What are the questions to which those observations are the beginnings of answers? I guess one is, "is it easy to earn outsized rents by selling to the gullible?". I presume most of the new sites make very little money. Whether they also attract away swoopo's customers or otherwise reduce swoopo's rents remains to be seen.

A new (job market) paper by Edward Augenblick at Stanford suggests that the already-established penny auctions may not disappear in the blink of an eye: Pay-As-You-Go: Theoretical and Empirical Analysis of a New Auction Format

He finds Swoopo to be quite profitable, and the abstract concludes:
"Finally, I attempt to address the long-term prospects of the market for these auctions. Using high frequency auction supply and user data, I estimate the current and optimal supply of auctions for a given number of users. This analysis suggests that the structure of the auction creates barriers to quickly developing a large userbase, allowing the most-established competitor to continue making large profits in the medium-term. This analysis is supported by auction-level data from five competitors. "

HT Eduardo Azevedo and Muriel Niederle