Abstract: This paper
presents our experiment on penny auction theory. Thirty-two participants
participated in our experiment and the experiment took place in 2015. We have
discovered that the theory of penny auction that stipulates that auctioneers
will lose if buyers collaborate with each other is true. There are several
strategies for penny auction. One of the strategies is to not involving in
penny auction. Another strategies are to be aggressive in order to win a penny
auction or bidders can collaborate with each other in order to make auctioneers
lose.
Introduction
Penny auction is an auction where bidders pay certain
amounts of cash to put their bids. Penny auctions differ from English auctions
in the sense that in penny auctions, bidders have to pay bid fees for each
price increment. Bidders in English auctions normally do not have to pay for
bid fees and only the winners have to pay for the final prices. However, in
penny auctions, all bidders have to pay whether s/he wins the auctions or not.
The number of bidders in an English auction normally will decrease as the price
increases, but the number of bidders in a penny auction does not necessary
decreases as the price increases. In penny auction, a product sells at a
starting price of zero dollars and bidders need to buy the bid right in order
to place their bids. The right to bid costs differently from one site to
another site and each bid increases the price by one cent and reset the
countdown time to one minute or 30 seconds or even 15 seconds, depending on the
auctioneers. Once the time reaches zero, the last bidder is declared the winner
and can purchase the object at the final price.
Penny auction does not have a very old history. It all
started back in Germany with a German company called Telebid (DealDash, 2014) . It began as series of
advertisements on TV with items that could be “bid” on by calling in on the
phone line. Then later, it expanded to the Internet with Swoopo.com in 2008 (DealDash, 2014) . Since then, many penny auction
web sites have been created and many of them were closed down or acquired by
other companies. For example, Swoopo.com was acquired by one of the US firms.
Penny
auction are interesting for several reasons. First, there is a heated debate
about whether penny auction is a gambling or not. Some people suggest policymakers
to ban penny auctions as they
consider a penny auction as a gambling or a lottery while others especially
auctioneers, have come up with various reasons to show that penny auction is
not a gambling or a lottery. Second, high revenues in penny auctions cannot be
explained fully by rational behavior; therefore, penny auctions provide
valuable empirical evidence for testing theories in behavioral economics (Hinnosaar, 2014) . Third, penny auctions are very
popular in real life and not many people understand about their properties and
the theory behind it.
We have conducted an
experiment on penny auction and we would like to present the results of our
experiment in this paper. The main purpose of our experiment was to test the
theories of penny auction. We used an instant camera as an item for our auction
and we conducted our experiment at our own institution, Zaman University.
Literature Review
There have been many papers written about the penny auctions.
One of them is written by Wang and Xu on “Selling a dollar for more than a
dollar? Evidence from online penny auctions.” The authors found that penny
auctions cannot sustain excessive profits without attracting new customers (Zhongmi Wang, Minbo Xu) . They found
this by analyzing the entrance and the exit time of the bidders from the penny
auction site BigDeal.com. They found that only a small percentage of bidders
who are sophisticated experienced bidders make significant profits compared to
others.
value of the auctioned good (Augenblick, 2011) . He argues that this is mostly
because of naïve sunk cost fallacy. In addition, he discusses about the
strategies in penny auctions such as zero-profit strategy, which is not joining
the auction at all, and aggressive bidding strategy.
Another useful paper about penny auctions is written by
Caldara on “Bidding behavior in pay-to-bid auctions: an experimental study.” In
his paper he explains that auction websites make large profits through mistakes
and inexperience of bidders (Caldara, 2012) . In addition to that, he states
that a combination of learning and the long-term attrition of less successful
participants threaten the mechanism to generate revenues.
The uniqueness of our experiment
There are several differences between our penny
auction and a regular penny auction. Penny auction usually takes place online,
but we conducted our penny auction on ground or at a real site. In this way,
penny auction bidders normally need to use Internet from buying the right to
bid to placing a bid. However, for our experiment, participants needed to buy
the right to bid from us directly and they needed to be present at the auction
site in order to place a bid. Some penny auction websites offer an automated
bidding tool for bidders to use, so bidders can still place their bids while
they are away from their computer. For our penny auction, participants could
not go anywhere while the bidding was taking place and they wanted to keep bidding;
otherwise, they faced a huge risk of losing the auction. This is another
difference between our penny auction and a regular penny auction. Several penny auction sites also use software
“bots” that automatically outbid people as the countdown clock reaches zero,
making it impossible for bidders to win items at reasonable price. Similarly,
some penny auctioneers have their friends bid in
order drive up prices and prevent consumers from winning. We as auctioneers did not manipulate the auction
as some penny auctioneers do. The process of our countdown time was also
different from the majority of penny auction sites. Typically, if
for a same bid, there are several bidders, the offers will be placed at once
back-to-back. The price of the product and the countdown will increase for each
offer placed. For
our experiment, when there were several bidders bid at the same time, we increased
the price to several folds and the countdown time would reset to 30 seconds and
later 15 seconds only, not to several folds as the price. Finally, there was a psychological test that
measures risk level of players in our penny auction. None of the penny
auctioneers have ever asked bidders to do psychological test before.
Hypothesis
As
aforementioned, the main purpose of our experiment is to test the theory of
penny auction. Penny auction theory states that an aggressive participant will
win the auction. This is our first hypothesis. The theory also asserts that
auctioneers will lose if buyers collaborate with each other. This is our second
hypothesis.
Experimental design
In order to conduct this experiment, first of all, we needed
to find participants. We made an announcement about our penny auction two weeks
before the auction date through Facebook, Evenbrite, and pamphlets. In order to
participate in our penny auction, participants needed to buy a registration
ticket from us at a price of $2. We
aimed to sell 50 tickets in order to break even before the auction took place;
however, we were able to sell 45 tickets only. Most of our buyers were students
from various majors at Zaman University. The registration ticket can be
exchanged for four bidding rights on the auction day.
Before the auction started, we asked our participant to do a
psychological test in order to know their risk behavior for the purpose to test
our first hypothesis. Once they completed the psychological test, we provided
each of them four bid right cards and for each bid s/he placed, we collected
back a card from him/her. If s/he were running out of the right to bid and s/he
still wanted to continue to bid, s/he could buy an additional right to bid at a
cost of 0.5$ each. S/he could buy the bid right as many as s/he wanted. A bid
summited by the participant constituted willingness of him/her to buy the
camera at the posted price.
The price of the camera started at zero and each bid
increased the price of the camera by one cent (100 cents = $1 USD). Initially,
each new bid reset the remaining time for an auction up to 30 seconds. Later,
we observed that our participants waited until the last 10 seconds to place
their bids, so we decided to reduce the countdown time to 15 seconds in order
to save time. The price of the product
would increase for each offer placed and the remaining time for an auction would
reset up to 30 seconds and later 15 seconds. If for a same bid, several bidders
bade at the same time, the offers would be placed at once back-to-back. For instance,
three participants bade for the camera at the price of 30 cents, the price of
the camera would increase to 33 cents and the last person who raised his/her
hand would be the last one who was to bid. The auction ended when the remaining
time reached zero. The
participant who was the last to bid then won the auction.
A
contract to purchase the item did not come into effect until the auction was
finished. Winning the auction did not
mean the participant who was the last to bid won the auction item: It meant that s/he
won the right to buy the item at the final price. After the auction
had ended, the last bidder “winner” should has paid the total price of the
Camera.
Data
and Result
We were able to sell 45 tickets and there were 32 people who
attend the auction. Before we began the auction, we asked our participants to conduct
a psychological test. We conducted this test in order to know our participants’
risk behavior for the hope to test/confirm our first hypothesis. We classified
our participants’ risk behavior into four categories: defensive, conservative,
moderate risk, and moderate aggressive. The majority of our participants fell
into the moderate risk category, followed by the moderate aggressive category
(Figure 1). Only 2 of our participants were conservative risk takers, while 1
of our participants was a defensive risk taker. According to our first
hypothesis, we would expect one of the moderate risk aggressive takers to win
the auction if there was no collaboration among the buyers.
The auction lasted for about an hour and the
camera was sold at a final price of 73 cents. The winner was Sophanha, who is a
conservative risk taker. Based on our observations, he won this auction because
he colluded with his peers. In this respect, he collaborated with other participants in the
auction namely: Mey, Mouy Eang, and Chhayhak. They took turn to raise their
hands when the countdown time almost reach zero and it seemed like no one else
in the auction placed the next bid and the last bidder was not one of them. By
doing so, it made other bidders hard to win the auction because they had more
bidding rights as a whole group than any other participant in the auction.
Based on our second hypothesis, if buyers collaborate with each other,
auctioneer will lose. This experiment has confirmed our second hypothesis. We lost around $40 from the auction as
a result of our participants’ collaboration.
Strategies for Penny Auction
One of the strategies for penny auction is to not participate
in this type of auction. Based on the past experience, only a very small
percentage of participants made profits out of penny auction, while the
majority of penny auction participants lost money. Thus, it is wise for us to
not participate in penny auction as there is a low chance of making profits
from penny auction and the risk of losing money is very high.
However, if one really wants to participate in a penny
auction, the theory suggests him/her to be aggressive in order to win the
auction or he/she can collaborate with other participants to make auctioneer
lose. But after all, these two strategies cannot guarantee a 100% chance of
winning the penny auction. As a result, we strongly do not recommend people to
participate in a penny auction.
Conclusion
The
purpose of our experiment is to test the theory of penny auction. We have
discovered that the penny auction theory that states that auctioneers will lose
if bidders collude with each other is true. The winner in our experiment
collaborated with three of his friends in order to win the penny auction. A
rational person will not participate in a penny auction. However, if one really
wants to participate in a penny auction, we would suggest him/her to be
aggressive or he/she can collaborate with other bidders in order to make the
auctioneers lose. One also needs to bear in mind that the aggressive strategy or collaborate still
cannot guarantee him/her win in a penny auction.
References
Zhongmi Wang, Minbo Xu. (n.d.). Selling a dollar for
more than a dollar? Evidence from online penny auctions.
Augenblick,
N. (2011). Consumer and producer behavior in the market for penny auctions:
theoretical and empirical analysis.
Caldara,
M. (2012). Bidding behavior in pay-to-bid auctions: an experimental study.
DealDash.
(2014, April 25). DealDash Tips. Retrieved from DealDash Tips:
http://www.dealdashtips.com/it-all-started-with-swoopo-a-brief-history-of-penny-auctions/
Hinnosaar,
T. (2014, 11). Penny Auctions . Retrieved from: http://toomas.hinnosaar.net/pennyauctions.pdf
No comments:
Post a Comment