2554
The Regulation of New Forms of Electronic Fund Transfers in Japan Focusing on Electronic Money
Electronic payment and legislation (p.596).
Tokyo: Yuhikak u.
19
 7KH¿UVWUHSRUW³7KHURXQGWDEOHFRQIHUHQFH
for Electronic Money and electronic payment”
(May 23, 1997) concluded that the Bill and
6LPLODU&HUWL¿FDWHV&RQWURO/DZZRXOGQRW
apply to IC-chip type (pp.39-40). Retrieved
March 15, 2007, from />p_mof/singikai/kinyusei/tosin/1a1201.htm
20
Osaka District Court Judgment, September
14, 2001.
21
FSA. (2004). Financial Advisory Agency No
Action Letter of April 20, 2004 to Barclay
Vouchers K.K. Retrieved March 15, 2007,
from />kaitou/001/001_06b.pdf and
FSA. (2004). Financial Advisory Agency
No Action Letter of July 9, 2004, to K.K.
Daiichi-Bussan. Retrieved March 15, 2007,
from />kaitou/001/001_08b.pdf
22
Iwahara, S. (2003). Electronic payment and
legislation (pp.596). Tokyo: Yuhikaku.
Mr. Fujiike, on the other hand, has suggested
that Electronic Money will fall within he
GH¿QLWLRQRI³H[FKDQJHWUDQVDFWLRQV´RQO\
if is has both General Cashability and ver-
satility together. This conclusion is realistic,

however, the reasons why both the elements
are necessary in order for Electronic Money
WREHFODVVL¿HGDV³H[FKDQJHWUDQVDFWLRQV´
are not clear. Fujiike, T. (2002). Public regu-
lations on settlement services by operating
FRPSDQLHVWKDWLVQRW¿QDQFLDOLQVWLWXWLRQV
Kinyu-houmu-jijo. 1631, pp.19-26, Tokyo:
Kinzai.
23
Nakazaki, T. (2007). Anti-money laundering
laws of Japan. In Anti-money laundering
International law and practice, pp. [uncer-
tain, to be published by the end of March].
London: Wiley & Sons.
24
Kubota, T. (2003). Legal issues related to
fund settlement systems (pp.179-208). Tokyo:
Kokusai-shoin.
25
 3OHDVHVHHWKHGH¿QLWLRQDWWKHVHFWLRQ$RI
³([LVWLQJ5HJXODWLRQVRQ(OHFWURQLF0RQH\²
3DUW,,,²&DVKDELOLW\²WKH'HSRVLW/DZ³
26
Kubota, T. (2003). Legal issues related to fund
settlement systems (p.191). Tokyo: Kokusai-
shoin.
27
Kubota, T. (2003). Legal issues related to fund
settlement systems (p.194). Tokyo: Kokusai-
shoin.

28
Concretely speaking, Item 2, Paragraph
6 , Article 7 of the new rule would be the
point.
29
Kubota, T. (2003). Legal issues related to fund
settlement systems (p.192). Tokyo: Kokusai-
shoin.
30
Kubota, T. (2003). Legal issues related to fund
settlement systems (p.193). Tokyo: Kokusai-
shoin.
31
E-gold is one of famous worldwide e-currency
companies and provides multinational online
payment service backed up by gold and other
metals. For more details, please visit the fol-
lowing URL: />32
Iwahara, S. (2005). The ideal future of
regulations on electronic money. In The First
Subgroup of the Study Group on the Financial
System of the Japanese Bankers Association
(Zenginkyo), E-money legislation (pp. 68-76).
Tokyo: Zenginkyo.
33
Maeda, Y. (2005). Ideal future legislation on
electronic money. In The First Subgroup of
the Study Group on the Financial System of
the Japanese Bankers Association (Zengin-
kyo), E-money legislation (pp. 1-5). Tokyo:

Zenginkyo.
34
 7KHJDPHZDV³5DJQDURN2QOLQH´RQHRIWKH
most popular multiplayer online role playing
games in Japan.
35
 )RUH[DPSOH³6HFRQG/LIH´www.secon-
dlife.comRI¿FLDOO\DQQRXQFHGVWDUWLQJWKHLU
service in Japanese from April 2007 early
2007.
36
FSA. (2004). Financial Advisory Agency No
Action Letter of April 20, 2004 to Barclay
2555
The Regulation of New Forms of Electronic Fund Transfers in Japan Focusing on Electronic Money
Vouchers K.K. Retrieved March 15, 2007,
from />kaitou/001/001_06b.pdf
37
Sugiura, N. (2003). Legal issues related
to point-based rewards programs. Kinyu-
zaisei-jijo, No.2561, p.41. Tokyo: Kinzai.
See also Sugiura, N., & Kataoka, Y. (2003).
Future Electronic Money and its legal infra-
structure (p.38). Retrieved March 15, 2007,
from />sion/2003/20030828-2.pdf
38
Maeda, S. (2005). Regulations on electronic
money. In The First Subgroup of the Study
Group on the Financial System of the Japanese
Bankers Association (Zenginkyo), E-money

legislation (pp.48-67). Tokyo: Zenginkyo.
39
Nomura Research Institute. (2006). Esti-
mates for nine major industries in Japan for
WKH  ¿VFDO \HDU Retrieved March 15,
2007, from ( />news/2006/060816.html
40
Nomura Research Institute. (2006). Business
currency in 2010. Tokyo: Toyo Keizai Inc.
41
Mr. Nakazaki. (2007). Three articles are
listed as dealing with predictable legal is-
sues arising from economic transactions in
Second Life in one of the largest IT news
portal Web sites in Japan, ITMediaBiz Web
site ( as fol-
lows: Nakazaki, T. (2007). Will real money
trading be a legitimate business? Retrieved
March 15, 2007, from edia.
co.jp/bizid/articles/0701/26/news008.html;
Nakazaki, T. (2007).
Converting in-game cur-
rency in U.S. Dollars at “Second Life”— is
this violating the investment law? Retrieved
March 15, 2007, from edia.
co.jp/bizid/articles/0702/15/news109.html;
Nakazaki, T. (2007).
Will a game contest of-
fering big money prizes in the virtual world
constitute an illegal gamble? Retrieved

March 15, 2007, from edia.
co.jp/bizid/articles/0703/16/news046.html
42
Five articles are listed as follows: Lemley, M.
A. (2002). Place and cyberspace. California
Law Review, 91, pp.521-549;
Lastowka, F. G., & Hunter, D. The laws of
the virtual worlds. California Law Review,
forthcoming;
Wu, T. (2003). When code isn’t law.
Virginia
Law Review, 89, pp.101-170;
G r i m m el m a n n , J. T. L . ( 2 0 0 4) . V i r t u a l w o r l d s
as comparative law. New York Law School
Law Review, 47, pp.147-184;
 )DLU¿HOG-9LUWXDOSURSHUW\%RVWRQ
University Law Review, 85, pp.1047-1102.
43
FSA. (2007). March 2007 newsletter in
English (pp.23-25). Retrieved March 15,
2007, from />ter/2007/02.pdf
This work was previously published in Cyberlaw for Global E-business: Finance, Payments, and Dispute Resolution, edited
by T. Kubota, pp. 142-167, copyright 2008 by Information Science Reference (an imprint of IGI Global).
2556
Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
Chapter 8.12
Pricing Strategy of Online
Knowledge Market:
The Analysis of Google Answers
Zuopeng (Justin) Zhang

Eastern New Mexico University, USA
Sajjad M. Jasimuddin
University of Wales at Aberystwyth, UK
ABSTRACT
This article addresses the different levels of
pricing strategies for an online knowledge mar-
ket. Based on the best practice from Google
Answers, an online knowledge market is mod-
eled as a marketplace where consumers ask and
researchers answer questions to make knowledge
transactions. Consumers optimally price their
TXH V W LRQ V W RREW DL QD Q VZH U VDQG D¿ U PPDL QW DL Q V
the online knowledge market by determining the
optimal price allocation to researchers. Our study
LGHQWL¿HVWZRW\SHVRIFRQVXPHUVVSLQRIIDQG
mainstream, based on whether additional utilities
will be derived from the market. In addition, we
LQYHVWLJDWH KRZWKH ¿UP FDQXVH PLQLPDO DQG
maximal posting prices to regulate the knowl-
edge market.
INTRODUCTION: ONLINE
KNOWLEDGE MARKET
Recent years have seen an enormous growth and
development of e-business. Internet technologies
have revolutionized the shopping behaviors of
FRQVXPHUVDQGWKHZD\V¿UPVDUHGRLQJEXVLQHVV
Firms launch electronic storefronts to advertise
their products and attract consumers to shop on-
line. According to the Global e-Commerce Report
by the Taylor Nelson Sofres (TNS) in 2002, 28%

of global Internet users and 32% of U.S. Internet
users have shopped online already or plan to do
so. The TNS report also indicates that U.S. is the
nation with the greatest proportion of Internet
users in the world who are engaged in online
shopping. Although online security remains the
biggest single concern for those Internet users
2557
Pricing Strategy of Online Knowledge Market
who have not yet shopped online, the undispu-
table fact is that more and more people are using
the Internet as the medium to browse product
LQIRUPDWLRQ¿UVWDQGWKHQPDNHSXUFKDVHV7KH
development of Internet technologies and the
proliferation of electronic intermediaries make
RQOLQHVKRSSLQJVLJQL¿FDQWO\HDV\EHFDXVHRQOLQH
shopping can save time and transportation costs
DVZHOODVSURYLGHFXVWRPHUVDGGLWLRQDOEHQH¿WV
such as product reviews, comparison of similar
products, best price search, and other advantages
ZKLFKDUHGLI¿FXOWWRREWDLQZKHQWKH\VKRSLQ
local stores. For example, Half.com and Ebay.com
provide online marketplaces for people around the
world to meet and make transactions, which has
VLJQL¿FDQWO\UHPRYHGRUUHOD[HGWKHWUDGLWLRQDO
constraints of time and location.
Internal knowledge markets have been found
in every organization (Matson, Patiath, & Shavers,
2003) in which knowledge experts and knowledge
seekers interact electronically to exchange their

knowledge. In a similar vein, recent years have
also seen the steady growth of online knowledge
markets specializing in various domain knowledge
and pricing systems. There are various online
knowledge markets available in the economy, such
as Intota.com, InfoRocket.com, Kasamba.com,
Knexa.com, Keen.com, eBrainx.com, Liveadvice.
com, Allexperts.com, and Swapsmarts.com. For
example, Intota Expert Knowledge Services
(www.intota.com) is a specialized service for
science and engineering, materials science, in-
dustry and technology, and business question
answering. Customers select an expert and contact
him(her) directly. Kasamba (www.kasamba.com)
KDVH[SHUWVLQD¿HOGZKRPFXVWRPHUVFRQWDFW
directly with their questions and a bid price.
Keen (www.keen.com) and LiveAdvice (www.
liveadvice.com) provide services to customers
who bid their questions to experts and will be
called back on the phone. Customers then pay
by the minute. Allexperts (www.allexperts.com)
RIIHUVFRQ¿GHQWLDOVHUYLFHVDQGXVHVGLUHFWHPDLO
to experts for customers’ questions. SwapSmarts
(www.swapsmarts.com) allows users to choose
prices for their posted questions. These business
practices provide classic examples for research
on external knowledge markets.
Google Answers (answer.google.com) is the
online market place similar to Half.com in the
sense that potential buyers and sellers can meet to

transact electronically. However, in contrast to the
traditional transaction of physical goods, buyers
VHWWKHLUSULFHVIRUWKH³JRRGV´WKHNQRZOHGJH
that they want to purchase, and sellers choose
from available offers from buyers to make trans-
actions. At Google Answers, customers can post
their questions and set a price between $2 and
$200. Researchers hired by Google browse all
the posted questions and decide whether or not
to answer the questions based on their own valu-
ations. A question can only be answered by one
researcher; once the answer is complete, 75% of
the price for the question will go to the researcher,
and the other 25% will remain to Google for its
maintenance fee.
Given the growing popularity of using online
knowledge markets to acquire knowledge, it is
naturally interesting to understand the working
mechanism behind the online knowledge market.
6SHFL¿FDOO\ZKDWDUHWKHRSWLPDOGHFLVLRQVIRU
D¿UPWRPDLQWDLQDQRQOLQHNQRZOHGJHPDUNHW"
How should a consumer optimally price his(her)
questions? When will a knowledge researcher
choose to answer questions posted by request-
ors?
Different levels of pricing strategies are inves-
tigated in this article. First, a consumer’s pricing
strategy is analyzed, and two types of consumers
are discovered on the knowledge market, spin-off
and mainstream consumers, based on whether

additional utilities are derived from the knowl-
edge market. Second, the reasons of specifying
minimal and maximal posting prices for the
NQRZOHGJHPDUNHWDUHLQYHVWLJDWHGWKH¿UPPD\
e l i m i n a t e s o me s pi n - o f f c o n s u m e r s b y d e s i g n a t i n g
DPLQLPDOSRVWLQJSULFHDQGLQFUHDVHLWVSUR¿WE\
mandating a maximal posting price. Third, the
2558
Pricing Strategy of Online Knowledge Market
optimal pricing strategy to researchers, or the
proportion allocated to researchers, is studied
by comparing the effects on market structure,
transaction price, and probability of questions
being answered when the allocation changes.
The article proceeds as follows. The next
section reviews relevant literature from the per-
spectives on knowledge market. The third section
outlines an analytical model of online knowledge
markets. The fourth section details our analysis
and discussion. The last section concludes the
article.
LITERATURE
The principle of knowledge market has been
recently applied to facilitate knowledge transfer
within an organization, which has also gained
growing attention in research. Davenport and
Prusak (1998) illustrate the concept of internal
knowledge market within organizations, and they
propose to employ the necessary IT support as
well as the indispensable incentives to build an

effective internal knowledge market for knowl-
edge transfer. Desouza, Awazu, Yamakawa, and
8PH]DZDGH¿QHNQRZOHGJHPDUNHWDV³DQ
environment where buyers and sellers can trade
WKHLUNQRZKRZZLWKLQGH¿QHGSULFLQJDQGWUDG-
ing rules”. Following the initial idea of knowledge
market within an organization, Ba, Stallaert, and
Whinston (2001) demonstrate that knowledge
c o m p on e nt s c a n b e o p t i m a l l y t r a d e d w it h a G r o ve -
Clarke-like mechanism within different bundles
LQDQLQWHUQDORUJDQL]DWLRQPDUNHWVRWKDWD¿UP
can optimally choose the knowledge bundles for
investment. While conducting a research, Desouza
et al. (2005) develop a mathematical analytics
to show the viability of the market mechanism
for knowledge management in organizations.
Mueller, Spiliopoulou, and Lenz (2002) formally
consider the electronic marketplace as an approach
to sharing knowledge assets. The characteristics
of knowledge as tradable goods to be transacted
on the e-marketplace are investigated within two
types of frameworks: the pricing system and the
quality evaluation method. In addition, internal
knowledge market is compared with the inter-
organizational knowledge market from various
perspectives in this study. Drawing from the
lessons from mini-cases, Desouza and Awazu
GH¿QHWKHQHFHVVDU\FRPSRQHQWVRIDQ
internal knowledge market and outline several
important caveats in association with econom-

ics literature when devising the market: market
of lemons, chicken-and-egg predicament, black
markets, and advertising strategies.
Several researchers have investigated the
online marketplace. Most notably, Bakos (1997)
regards an online marketplace as a special type
of electronic marketplace and proposes that
HOHFWURQLFPDUNHWSODFHVUHGXFHLQHI¿FLHQFLHVE\
lowering buyers’ cost to acquire information about
sellers’ prices and product offerings. Bakos (1998)
also adds that electronic marketplaces serve the
role of matching buyers and sellers and facilitat-
ing transactions, where increasing differentia-
tions and lower cost of product information can
be observed. Nevertheless, there lacks research
on online knowledge markets, especially how
they function and with what types of consum-
ers they deal. Several studies (Edelman, 2004;
Kenney, McGovern, Martinez, & Heidig, 2003)
have recently attempted to study the impact and
implications of an online knowledge market from
the inspirations of Google Answers. In a study
conducted by Cornell University Library (Ken-
ney et al., 2003), its digital reference services
are compared with those of Google Answers,
which provides an opportunity for librarians to
borrow valuable insights from Google’s approach
to service development and delivery. From the
behavioral perspective, Edelman (2004) analyzes
all auctions since the inception of the Google

$QVZHUV VHUYLFH $FFRUGLQJ WR KLV ¿QGLQJV
there exist certain notable trends in research-
ers’ behaviors. More experienced researchers
provide answers with more added value asked
2559
Pricing Strategy of Online Knowledge Market
by the requestors, receiving higher rankings as a
result. A researcher’s rate of earnings increases in
his(her) experience. In addition, a researcher who
only answers questions in particular categories
will provide answers of higher quality but will
earn less.
However, these studies have not investigated
the working mechanism behind an online knowl-
edge market, that is, the pricing strategies of
FRQVXPHUVDQGWKH¿UPZKRPDLQWDLQVWKHRQOLQH
knowledge market. Our research addresses this
JDSE\VSHFL¿FDOO\LQYHVWLJDWLQJWKHPRGHODSSOLHG
by Google Answers and makes contributions to the
literature by identifying two types of consumers
on the knowledge market. Our assertion is dif-
ferent from the prior literature on horizontally-
differentiated markets (Anderson & Palma, 1992;
Bockem, 1994; Cohen & Whang, 1997; Lancaster,
1990), where consumers are assumed to have
different preferences on products with respect
to quality, color, or other characteristics. Hence,
WKH¿UPVRQWKHPDUNHWPD\EHEHWWHURIIWRRIIHU
SURGXFWVZLWKGLIIHUHQWVSHFL¿FDWLRQVLPSOHPHQW
discriminating pricing policies, or collaborate with

competitors so as to manipulate the market. Our
research does not make assumptions on consum-
ers’ preferences, and they are only different with
UHVS H F WWRW KHL U VHD UFK LQJV N L O O V:H¿ Q GWKDW HYHQ
the consumers with the same searching skills can
belong to different types, either the spin-off or
the mainstream type, depending on whether they
are utilizing the knowledge market seriously to
derive additional utilities. The detailed analysis
and discussion start from the next section.
THE MODEL OF ONLINE
KNOWLEDGE MARKET
This section outlines our analytical model and
shows some preliminary results for further
DQDO\VLV:H¿UVWGHVFULEHWKHPRGHOZLWKVRPH
assumptions and then formulate the maximization
problem of each stakeholder on the knowledge
market.
:HFRQVLGHUD¿UPWKDWKLUHVNQRZOHGJHUH-
searchers to answer questions posted by customers
on the online knowledge market. The competi-
tion for people to be recruited as a researcher
E\WKH¿UPLV¿HUFHVRWKH¿UPFDQDOZD\V¿QG
the most intelligent and capable individuals to
work as researchers for the knowledge market.
Customers come to the knowledge market, post
their questions, and set the prices p for them.
Since customers do not know who will eventually
answer their questions before setting the prices,
they may always price their questions as suggested

by the guidelines of Google Answers:
³ 7 KH P R U HU HVH D UFK U H TX L UH G W R¿ QG D QD Q V Z H U
the higher the price you should set for your ques-
tion Setting a price too low to compensate for
the time required may result in your question not
receiving an answer. The more you are prepared
to pay, the more likely your question is to get
answered quickly.”
When a question is posted on the knowledge
market, all the researchers will have the equal
opportunity to answer it, depending on who no-
WLFHVLW¿UVWEHFDXVHRIWKHORFNLQJPHFKDQLVP
HPSOR\HGE\WKH¿UP%DVLFDOO\DQRSHQTXHVWLRQ
can be locked by a researcher for a certain period
of time so that only one researcher is allowed
to work on it at a time. Consequently, either the
question is answered and not available anymore,
or it becomes open again to all the researchers
if the one who locked it was not able to answer
it. Therefore, researchers try to catch questions
within their knowledge domains as quickly as
they can from the knowledge market. However, a
researcher cannot negotiate price with customers,
EXWZLOO LQVSHFW WKHGLI¿FXOW\RIHDFKTXHVWLRQ
and the price tag attached to it to decide whether
or not to answer the question. A question m has
its type q
m
RIGLI¿FXOW\ZKLFKLVGLVWULEXWHGZLWK
the probability function H(q

m
) between 0 and 1.
$PRUHGLI¿FXOWTXHVWLRQKLJKHUq
m
) requires a
2560
Pricing Strategy of Online Knowledge Market
researcher to have a higher knowledge level or
exert more effort to answer it.
Therefore, a researcher tries to maximize
his(her) total surplus by determining whether
or not to answer the questions he(she) observes
RQWKHPDUNHW)RUDVSHFL¿FTXHVWLRQm priced
at p, if he(she) answers it, a researcher will get a
net payoff as
S
r
= Dp – T(p, k
i
)  q
m
(1)
where DLVWKHSURSRUWLRQDOORFDWHGE\WKH¿UP
for answering each question with price p and T(p,
k
i
) is the disutility a researcher with knowledge
level k
i
will incur by answering this question,

including the effort cost and the risk of getting
a bad evaluation. The knowledge level k
i
is not
observable, but can be inferred from a known
distribution G(k
i
). We assume that T(p, k
i
) is an
increasing function in p and a decreasing func-
tion in k
i
with T(0, k
i
) = 0. It is also assumed that

2
T
2
i
k
DQG
2
Tp
2
&RQVHTXHQWO\WKH
following lemma can be obtained.
Lemma 1. Only those researchers with knowledge
levels k

i

ˆ
i
k
will answer the question m with a
price p, where
ˆ
i
k
is the threshold knowledge level
such that;
D
p = T(p,
ˆ
i
k
)  q
m
. (2)
Proof: A researcher will only answer a question
when his(her) total payoff from this question is
positive. For a given question m with price p, only
those researchers with k
i

ˆ
i
k
will have positive

payoffs, where Dp = T(p,
ˆ
i
k
)  q
m
 Ŷ
7 KL V O H P P DL G H QW L ¿ H VW KH WK UH VK R O G N Q RZOH G J H 
level
ˆ
i
k
; for those researchers with a knowledge
level higher than or equal to
ˆ
i
k
, they will prefer
to answer the question m with a price p since
they will get a non-negative surplus, whereas
for those lower than
ˆ
i
k
, they will not answer the
question. From another perspective, if a question
mLVPRUHGLI¿FXOWq
m
is larger), it will require a
resea rcher t o be m or e k n owle dge able to be able to

answer it because the threshold knowledge level
ˆ
i
k
increases in q
m
.
D proportion of price allocated to researchers
c()
FRVWRI¿QGLQJWKHDQVZHUZLWKRWKHUPHWKRGV
k
i
researcher i’s knowledge level
F(s
j
)
cumulative distribution of consumer skill level
G(k
j
)
cumulative distribution of knowledge level
H(q
m
)
FXPXODWLYHGLVWULEXWLRQRITXHVWLRQGLI¿FXOW\
p
price of a question set by knowledge consumers
p
minimal posting price on the knowledge market
ˆ

p
maximal posting price on the knowledge market
U() probability of yielding an additional utility
q
m
GLI¿FXOW\RIDTXHVWLRQm
s
j
level of search skills for consumer j
T() disutility of a researcher answering a question
Table 1. Summary of notation
2561
Pricing Strategy of Online Knowledge Market
In addition, the above lemma assumes that a
researcher’s disutility in answering a question is
positively related to its price, which stems from
WKHUHSXWDWLRQUDWLQJV\VWHPDGRSWHGE\WKH¿UP
for the knowledge market. In general, questions
with a higher price are more vulnerable to getting
a lower rate from their originators for the quality
of the answer, because those questions are gener-
DOO\PRUHGLI¿FXOWDQGFXVWRPHUVDUHOHVVHDVLO\
pleased for a larger amount of payment. In this
sense, although questions with high price tags
VHHPPRUHSUR¿WDEOHWKH\DUHPRUHVXVFHSWLEOH
to bad evaluations; therefore, researchers from a
long-term perspective may not be willing to risk
their reputations to answer those expensive ques-
tions. In contrast, questions with low price tags,
DOWKRXJKWKH\DUH OHVVSUR¿WDEOHPD\EHPRUH

popular to researchers because they may help build
the long-term reputations on the market.
Following the above lemma, a question with
a price p to be answered is
1
()
i
ii
k
gk dk
³
ˆ
,
where g(k
i
) is the probability density function of
researchers’ knowledge levels and we assume that
k
i
is uniformly distributed in [0, 1]. From the above
discussion, it can be seen that the probability of a
question to be answered decreases with its price,
whereas the quality of the answer increases in its
price because it will be answered by a researcher
with a higher knowledge level. If the question m
is not answered, a customer j will incur a cost
c(s
j
, q
m

) for exerting effort to get the answer by
himself(herself) or from other means in order to
enjoy the utility u, where s
j
describes the ability of
the customer jLQ¿QGLQJWKHDQVZHUZLWKRXWWKH
help of Google Answers. For instance, customers
m a y u s e t h e G o o g l e s e a r c h e n g i n e t o s e a r c h fo r t he
relevant documents by themselves. The amount
of time they spend varies with their skills in us-
ing the search engine. Typically, a more skillful
consumer (a higher s
j
) incurs less cost c(s
j
, q
m
)for
¿ QG L Q J W KHD QV ZH U W RW KH V D PH T X H VW L R Q , IK LV  KH U
question mLV¿ QDOO\D QVZHUHGW KHFXVWRPHUZKR
sets the price p for his(her) question m may get
additional utility 'u(q
m
) with the probability U(p),
which increases in the price. The probability U(p)
of obtaining additional utility being an increasing
IXQFWLRQRIWKHSULFHFDQEHMXVWL¿HGIURPWKHIRO-
lowing two aspects. First, questions with higher
prices will be answered by researchers with higher
knowledge levels; therefore, consumers may get

better answers if they set higher prices for their
questions. Second, as shown later, a question with
a higher price implies that its originator is more
knowledgeable or values his(her) question more;
hence, researchers will try to provide a better
answer in order to be well evaluated.
A consumer’s level of searching skills s
j
is
XQNQRZQWRUHVHDUFKHUVDQGWKH¿UPDQGFDQ
only be perceived from a probability distribution
function f(s
j
). Therefore, a customer with the skill
level s
j
maximizes his(her) expected surplus,
[()()]
cm
upuqp  ' 
1
() [ ( )]
i
ii jm
k
gk dk u cs q
³
1
1()
i

ii
k
gk dk
ªº
«»
«»
¬¼

³
(3)
E\¿QGLQJDEHVWSULFHp for his(her) question on
the knowledge market. Given the questions posted
RQWKHNQRZOHGJHPDUNHWWKH¿UPPD[LPL]HVLWV
expected payoff from these questions,
1
(1 ) ( )
mj
i
qs i i
k
EE p gkdk
ªº
«»
«»
¬¼
 
³
(4)
by specifying the appropriate allocation D. The
allocation D, the actual proportion of payment

researchers will get for answering each question,
c a n b e r eg a r d e d a s t h e i n c e nt iv e t o i n d uc e r e s e a r c h -
ers to answer questions because they get more
out of answering each question for a larger D. In
addition, it can be inferred from Lemma 1 that a
larger D increases the success rate of knowledge
2562
Pricing Strategy of Online Knowledge Market
³WUDQVDFWLRQ´TXHVWLRQVEHLQJDQVZHUHGVLQFH
the threshold knowledge level
ˆ
i
k
decreases in
D. Intuitively, if a researcher can get more from
DQVZHULQJHDFKTXHVWLRQRULIWKH¿UPDVVLJQVD
higher D), he(she) will be willing to answer those
questions with lower prices because he(she) will
be able to extract positive surpluses from these
questions with a higher allocation D. Therefore,
each researcher will be willing to answer more
questions with a larger allocation D, which implies
that a question will become lucrative to more
researchers, increasing its probability of being
answered. On the negative side, increasing the
allocation D may jeopardize the quality of answers
WRHDFKTXHVWLRQDQGGHFUHDVHWKH¿UP¶VQHWSUR¿W
+HQFHWKH¿UPKDVWREDODQFHWKHVHWUDGHRIIVWR
choose a best allocation to researchers.
ANALYSIS AND DISCUSSION

This section details our analysis and discussion on
various pricing strategies of an online knowledge
market. Beginning with the individual consumer’s
best pricing strategy, we identify two types of
consumers on the knowledge market, spin-off
information consumers and mainstream knowl-
edge consumers, with regard to whether addi-
tional utilities can be derived from the knowledge
market. Based upon this differentiation, we then
discuss the purpose of specifying minimal and
PD[LPDOSRVWLQJSULFHVDQG¿QDOO\DQDO\]HWKH
¿UP¶VRSWLPDODOORFDWLRQWRUHVHDUFKHUV
Consumers’ Pricing Strategy
A knowledge consumer’s best pricing strategy
is investigated in this subsection, leading to the
differentiation of two types of consumers on the
knowledge market.
First of all, a knowledge consumer will always
set his(her) best price for a posted question on
the market. Assuming a uniform distribution for
G(k
i
), we can rewrite a consumer j’s total surplus
from Equation 3 as
c

(( ))[() ( )](1)
jm jm
i
ucsq pucsq p

k
  '
where the second term is contingent on
ˆ
i
k
, the
lowest knowledge level required to answer the
TXHVWLRQ ZKLFKÀXFWXDWHV ZLWK WKH FXVWRPHU¶V
price p according to Lemma 1. Hence, the cus-
tomer can price his(her) question according to
the following lemma.
Lemma 2: A knowledge consumer will set his(her)
question with the optimal price p
*
where
() 1
() ( )
jm
pu
pucsq p


c
'
'  
1
i
i
p

k
k
ww


(5)
and Up
*
'u p
*
– c(s
j
, q
m
).
Proof:7KH¿UVWRUGHUFRQGLWLRQRIS
c
with respect
to p is
(1 –
ˆ
i
k
)(U'(p) 'u – 1) – (U(p) 'u + c(s
j
, q
m
)
ˆ
i

k
p
w
w
=
0
and the second order derivative is
2
2
c
p
w
w
(1 –
ˆ
i
k
)U''(p) 'u – (U(p) 'u +
c(s
j
, q
m
) – p
2
2
i
k
p
w
w

ˆ
< 0
if U(p) 'XS
*
– c(s
j
, q
m
) EHFDXVH
2
ˆ
i
k
S
2
> 0,
which can be derived from Lemma 1 and our
assumptions about a researcher’s disutility that

2
T/l
2
i
k
0 and 
2
Tp
2
0 Ŷ
L e m m a 2 s u g g e st s t h a t t he r e e x i s t s a b e s t p r i c e

p
*
IRUDFXVWRPHUWRSRVWDVSHFL¿FTXHVWLRQWR
maximize his(her) surplus because the LHS of
2563
Pricing Strategy of Online Knowledge Market
Equation (5) decreases and the RHS increases
in p. Therefore, an intersection always exists be-
tween the curves of LHS and RHS with respect
to the price p. For a consumer with a better skill
level (larger s
j
) or a larger additional utility, the
LHS increases, resulting in a larger p
*
. However,
the condition, U(p
*
) 'u p
*
– c (s
j
, q
m
), has to be
VDWLV¿HGZKLFKLPSOLHVWKDWDFXVWRPHUVKRXOG
always offer a price p
*
less than his(her) own cost
c(s

j
, q
m
RI¿QGLQJWKHDQVZHUXQOHVVWKHH[SHFWHG
additional utility U(p) 'u he(she) may derive from
WKHNQRZOHGJHPDUNHWFDQEHMXVWL¿HGE\RIIHU-
ing the price.
A knowledge consumer’s pricing strategy
critically depends on the additional utility 'u
he(she) may obtain from the knowledge market.
It can be inferred from Equation 5 that under
certain conditions, a knowledge consumer will
set a higher price for a question if the answer can
bring more additional utility 'u, which is stated
in the following proposition.
Proposition 1: A consumer will set a higher price
p
*
for his(her) question if he(she) can derive more
additional utility 'u from it and
() 1
() ( )
jm
p
ppcsq


c
!


(6)
Proof: The derivative of RHS of Equation 5 with
respect to 'u is zero. Hence, the optimal price
p
*
will increase with the additional utility 'u if
U'(p
*
) (c(s
j
, q
m
) – p
*
) + U(p
*
), the derivative of
LHS of Equation (5) with respect to
'u is greater
than zero, which is true if the above condition
KROGV Ŷ
The condition in the above proposition par-
allels that in Lemma 2, implying again that the
cost c(s
j
, q
m
) is a critical point for a consumer to
decide whether or not to increase his(her) price
when he(she) may derive more additional utility

from the answer of the question. The condition sug-
gests that when p is less than c(s
j
, q
m
), the condition
always holds, whereas when p is greater than c(s
j
,
q
m
), the condition may not hold. When the addi-
tional utility keeps increasing, the optimal price
increases as well until the critical point c(s
j
, q
m
),
beyond which point the consumer has to consider
whether keeping or increasing the optimal price
will increase the probability of obtaining the
additional utility so that the increased cost of
payment can be compensated.
In summary, when a consumer seriously seeks
answers from the knowledge market to derive
DGGLWLRQDOXWLOLWLHVKHVKHPD\¿QGDEHVWSULFH
to post his(her) question on the market under
certain general conditions. However, when the
additional utility 'u is very small, especially
when 'Xĺ0, a consumer’s pricing strategy will

change dramatically, which is presented in the
next proposition.
Proposition 2: When 'XĺRUQRDGGLWLRQDO
utility is derived from the answer of his(her) ques-
tion, a consumer will set the price at
p
for his(her)
question, where
p
is the minimal price required
to post a question on the market.
Proof: When 'XĺDNQRZOHGJHFRQVXPHU¶V
expected payoff is
S
c
= (u – c(s
j
, q
m
)) + [c(s
j
, q
m
) – p](1 –
ˆ
i
k
),
ZKHUHWKHVHFRQGWHUPWKHEHQH¿WGHULYHGIURP
the knowledge market, decreases in the price p.

Therefore, the optimal price is achieved as the
boundary solution, that is, the consumer will set
the price at
p
, the minimal price to post a question
RQWKHPDUNHW Ŷ
The phenomena described in the above propo-
sition can be truly observed from the knowledge
market maintained by Google Answers; about
15% of questions are priced at the minimal posting
price, $2. In such case, the consumers do not care
whether they will be able to derive any additional
utilities from the market, but regard the market as