The Internet, Policy & Politics Conferences

Oxford Internet Institute, University of Oxford

Melisande Cardona, Georgios Alaveras: Consumer Choice and Privacy Settings in the App Market

Melisande Cardona, JRC

Georgios Alaveras

More than 60% of European internet users access the internet via a smartphone (European Commission, 2015)and apps are shaping a large part of the digital platform economy the last few years. Mobile applications is one of the industries with the most rapid growth. In the end of 2010 there were 100,000 apps available in Google Play Store and around 300,000 for the Apple App Store. In 2015 there were more the 1,400,000 applications reported in each of the Google Play and Apple App stores. In addition to this, 100 billion downloads were reported in the Apple Store by the mid-2015. The two most popular apps (Whatsapp and Facebook) have each been downloaded more than 1 billion times on the Android platform only.

The majority of mobile applications in order to function properly require access (‘permissions’) to information that can be considered private from several users, such as the user's geo-location or her contacts. This can deliver benefits to consumers through enhanced services from geolocation, but can also be used for personalization or financial rewards in the form of free apps. At the same time smartphones usually carry more personal information than other IT devices (e.g. geolocation, contacts). In fact the amount of personal information stored on smartphones is regarded so valuable that it has become the issue of a public debate between the FBI and Apple on granting backdoor access to smartphones. Due to the huge amount of personal information stored on phones, the issue of granting access to this data to app providers and having knowledge about the use thereafter would be expected to be of interest to the users. The question arises whether and how these concerns about collecting private information from the smartphones prevail in consumer choices. We therefore study how permission requests of apps affect consumers in their usage and adoption decisions of apps.

89% of European internet users state that they avoid disclosing personal information (European Commission, 2015), but this does not necessarily have to lead to more reserved sharing of such in consumer behaviour. The privacy paradox describes this often observed phenomenon that attitudes and behaviour towards privacy diverge. In economic analysis this can simply be seen as the outcome of rationally weighing the benefits and concerns of revealing personal information against each other (Smith et al., 2011). Though the privacy paradox can also be explained by the uncertainty about its consequences especially when informational data is concerned (Acquisti et al., 2015).

As of yet, there is not much literature on the subject. Mobile applications are a new market that has matured very fast without a comparable quantity of research. Notable exceptions have been studies on innovation and value creation of apps (Bresnahan et al., 2014, Yin et al., 2014). The only study to our knowledge that combines the issue of apps with privacy is work by Kummer and Schulte (2015) which is based on aggregate number of downloads. They find that on the supply side, free apps have a strong tendency to ask for more privacy related information and that on the demand side number of downloads is negatively correlated to apps asking permission to privacy relevant information. We build on their research and use the same classification of privacy relevant permissions based on Sarma (2012). To study whether the privacy concerns enter the demand function, we use data from a panel obtained from GfK, which tracks app usage at the individual level from January to August 2015 in Germany. This is an interesting empirical setting to study consumption choices and the privacy paradox, because it is a country whose population shows above average concern when asked about its attitude towards privacy issues (European Commission, 2015). The information on different properties and the permission requests of the apps was retrieved from the Google Play Store in September 2015.

For the first regressions we use adoption and usage of the Android platform. Google Play store offers transparency to the extent that all privacy settings the app requests are listed before the app installation decision is confirmed and certain permission settings are flagged with a header (e.g. contacts, identity, or location). Nevertheless, consumers have no control to deselect certain permissions settings at purchase or afterwards, they have to accept the permission settings in the bundle in order to install and use the app. This privacy policy when purchasing apps on the Google Play store allows us to understand which permission settings are accepted when installing an app.

We estimate the demand for usage of approximately 10,000 different apps (log-lin fixed effects panel estimation) for nearly 5,800 consumers and the adoption decision on 70 more homogeneous ‘music-audio’apps (Conditional multinomial logit estimation). The first results indicate that while the total number of permissions is positively related to increased demand, flagged permission requests have a negative effect on both usage and adoption decisions. This is an indication that consumers reveal privacy concerns in their consumption behaviour and that Google Play Store’s presentation of the requests make a difference. We further find that age and gender matter, while education shows little difference regarding decisions concerning privacy relevant permissions.

In the next steps of the analysis we will compare adoption between the Apple and Android users, in order to explore how the difference in privacy policy affects usage and adoption. This will enable us to test the hypothesis whether privacy concerns slows adoption. Another step could be to distinguish adoption decisions regarding privacy preferences between big established apps and smaller apps which face more competition. Furthermore, we will identify privacy aware consumers through their media consumption and app-adoption and study how their demand differs from the remaining population.


Acquisti, Alessandro; Laura Brandimarte and George Loewenstein. 2015. ""Privacy and Human Behavior in the Age of Information."" Science, 347(6221), 509-14.

Bresnahan, Timothy; Jason P Davis and Pai-Ling Yin. 2014. ""Economic Value Creation in Mobile Applications"", Manuscript. European Commission. 2015. ""Cyber Security,"" Special Eurobarometer 423.

Kummer, Michael E and Patrick Schulte. 2015. ""Money and Privacy’Android Market Evidence"", Manuscript.

Sarma, Bhaskar Pratim; Ninghui Li; Chris Gates; Rahul Potharaju; Cristina Nita-Rotaru and Ian Molloy. 2012. ""Android Permissions: A Perspective Combining Risks and Benefits,"" Proceedings of the 17th ACM symposium on Access Control Models and Technologies. ACM, 13-22.

Smith, H Jeff; Tamara Dinev and Heng Xu. 2011. ""Information Privacy Research: An Interdisciplinary Review."" MIS quarterly, 35(4), 989-1016.

Yin, P. L.; J. P. Davis and Y. Muzyrya. 2014. ""Entrepreneurial Innovation: Killer Apps in the Iphone Ecosystem."" American Economic Review, 104(5), 255-59.

Melisande Cardona, Georgios Alaveras