The Value of Personal Information in the E-Commerce Market Toshiya Jitsuzumi1 and Teppei Koguchi2 1 Faculty of Economics, Kyushu University, Japan 2 Faculty of Informatics, Shizuoka University, Japan
Abstract The purpose of this analysis is to clarify the effect of pe rsonal information on switching costs in the Internet sh opping market. We empirically demonstrate the extent to which person al information drive up switching costs. We revealed that when users change Internet shopping sites, personal information registered on the site represe nt switching costs of the same magnitude as traditional switching costs.
Background: Internet shopping in Japan Growth of the Japanese B2C e-commerce market 8.6 % other Retail and service industries Source: Ministry of Economy, Trade and Industry
Background: Internet shopping in Japan Shares of Internet shopping sites in 2012 in Japan Source: Rakuten, Inc. fiscal 2012 Financial Results
Background: Personal information on Internet shopping In order to shop on Internet shopping sites, users must register and provide information. ◦ names, e-mail addresses, postal addresses, credit card numbers, etc. In addition, many Internet shopping sites provide the user’s viewing and b uying histories on the site to make shopping more convenient. If a user changes to another Internet shopping site, ◦ The personal information that has been registered and stored on previously used site is not transferred to the new site. The user must re-register his or her personal information. The personal information on previously used site may be continuously used for the business of prev iously site. ◦ Viewing and buying histories on previously used site can’t watch in the new site. The user can’t use wish list and recommendation function based on buying history. This point may represent a switching cost for users.
Background: Early studies Switching Cost; ◦ The psychological or economic costs incurred when customers switch from one good or se rvice to another. Traditional switching cost; familiarity, attachment, etc. to the service. ◦ Switching costs make consumer hard to switch different service. If high switching cost exists, it is possible to prevent price competition. Klemperer (1987) ◦ Analysis for competition between new entry brand and existing brand. Shy(2002) ◦ A model analysis of switching costs in the financial industry. Valletti and Cave (1998) ◦ Analysis the mobile phone market in the UK. Brynjolfsson and Smith (2000) ◦ In the e-commerce market, consumer confidence in the service provider becomes the factor of switching costs and justifies price differences.
Analysis framework: Scenario Hypothesis of scenario “Rakuten and Amazon will be integrated, and one of them will close.” Respondents recognize as switching costs for; ◦ Traditional switching cost familiarity and attachment to the site that will close. ◦ Switching cost associated with personal information the management of registered information (names, e-mail addresses, credit card numbers, etc.) on the site that will close. the migration of the viewing histories at the site. the migration of the buying histories at the site.
Analysis framework: Attributes and levels
Analysis framework: Conjoint analysis Probability function exp 'X i
' exp j g d X i
Utility function (Without shift parameter) (With shift parameter) U Amazon Amazon Amazon , D
, D ij Amazon i Amazon Rakuten i Rakuten U ( age freq purchase ) , D ij age i freq i purchase i Amazon i Amazon inf, Dinf , D , D i buy i buy view i view Rakuten Rakuten Rakuten
( age freq purchase ) , D
age i freq i purchase i Rakuten i Rakuten , compensat o i n compensation i ij inf ( age inf freq inf purchase ) inf, D age i freq i purchase i i inf buy buy buy
( age freq purchase ) , D age i freq i purchase i buy i buy view view view
Variables; age freq purchase D age i freq i purchase i view i view D means dummy variable
, compensat on i
compensat o i n i ij If DAmazon = 1, merging into Amazon If DRakuten = 1, merging into Rakuten If Dinf =1, deleting registered information Shift parameter; If Dbuy =1, carrying over buying history age = age If Dview =1, carrying over viewing history freq = purchase frequency during last year If Dcompensation =1, compensation for each situation (yen) purchase = average purchase price
Estimation: Results Without shift parameter With shift parameter
Estimation: Results; WTA
Conclusion Purchase histories or registered personal information represent switching costs of the same magnitude as traditional switching costs such as brand attachment or familiarity with the site. Viewing histories are not regarded as the factor of switching c osts. From the managerial perspective; ◦ It is effective to construct a system in which registered or stored persona l information cannot be used at different sites. ◦ Especially, for young people, it is important to apply the services, for ex ample reduced prices, to prevent from changing to different service prov iders.
From the perspective of government policies; ◦ It is necessary to analyze what types of personal information are register ed or stored on these sites. While some personal information become switching costs, others do not. ◦ If switching costs impede competition, we have to consider the policy th at makes possible transportation of personal information. The “midata” project (BIS in 2011). The goal of the “midata” project is for consumers to be able to access and use their personal and company data. This project would be able to solve the problem of switching costs associated with personal information and therefore promote more competition.