Payment Instruments, Payment Instruments, Payment Instruments, - PowerPoint PPT Presentation

Payment Instruments, Payment Instruments, Payment Instruments, Payment Instruments, Financial Privacy Financial Financial Financial Privacy Privacy Privacy and Online Purchases and Online Purchases and Online Purchases and Online Purchases Balgobin Y., Bounie D., Quinn M. and Waelbroeck P. Telecom ParisTech DNB Conference, April, 21-22 2016 De Nederlandsche Bank

1993 1993 1993 1993 De Nederlandsche Bank

2015 2015 2015 2015 De Nederlandsche Bank

Willingness to share data in France Willingness to share data in France Willingness to share data in France Willingness to share data in France (2009 (2009 (2009 (2009 – – – 2015) – 2015) 2015) 2015) The protection of personal data is a major concern in the digital economy. 5% do not want to share any data in 2009 and 21% in 2015 . 20 15 % of respondents 10 5 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Number of information 2009 2011 2013 2015 De Nederlandsche Bank

Financial privacy Financial privacy Financial privacy Financial privacy � Financial data are also concerned with privacy. Why ? � Online purchases are paid with cards that leave traces. � Traces are exploited by banks to offer rewards, financial services, etc., but also to grant credits (credit scores). � Privacy-minded consumers may refuse to shop online not to disclose data to banks and preserve a high credit score, to avoid to be tracked, or simply not to be targeted. � Consequence : privacy-minded consumers may reduce their online purchases as bank payment instruments (PI) are too costly in terms of privacy. De Nederlandsche Bank

Research questions Research questions Research questions Research questions � Solution : using a payment instrument (PI) that is not related to a banking account: a non-bank PI. � PayPal, private label cards, virtual currencies, etc. � Questions : � Does financial privacy affect online purchases? � Does the use of non-bank PI increase the purchases of privacy- minded consumers (with respect to others)? De Nederlandsche Bank

Paper (preliminary version) Paper (preliminary version) Paper (preliminary version) Paper (preliminary version) � A simple (IO) model. � Two regressions to test a model prediction. � Estimation results and preliminary conclusions : • The use of non-bank PI has a positive impact on the frequency of online purchases • The use of non-bank PI is driven by financial privacy as we control for payment convenience, risk, etc. • The use of non-bank PI by privacy-minded consumers increases their online purchases compared to others consumers. De Nederlandsche Bank

Literature review Literature review Literature review Literature review � Financial privacy and regulation of financial intermediaries. � Economics of privacy (Acquisti, 2015; Akther, 2012): privacy concerns affect negatively consumers’ online spending. � Economics of payment instruments (von Kalckreuth et al. (2014), Schuh and Shy (2016): anonymity explains adoption and use of instruments. � (Information sharing in credit markets). � Our paper is the first to directly tackle the issue of financial privacy in relation with payment instruments and online purchases. De Nederlandsche Bank

Survey Survey Survey Survey � Conducted in May 2015. � 1,000 French Internet users (15+) representative of the online population. � Objective: assess the level of trust of the Internet users in several online services (bank, administration, etc.). � Data on: • Purchases (frequency, average spending) and online activities. • Use of payment instruments. • Privacy concerns. • etc. De Nederlandsche Bank

Frequencies of online Frequencies of online Frequencies of online Frequencies of online p p p purchases urchases urchases urchases 81% of the respondents (n=811) have made at least 1 purchase during the last 12 months. 61% report to make more than one purchase per month. 60% 49% 50% Percentage of respondents 39% 40% 30% 20% 9% 10% 4% 0% Less often than once More than once per Once per week (N = 72) Several times per week per month (N = 316) month (N = 394) (N = 29) De Nederlandsche Bank

Payment instruments Payment instruments Payment instruments Payment instruments 95% declare using a payment card, and 36% PayPal. 100% 95% 90% 80% Percentage of respondents 70% 60% 11% gift cards + 50% 11% others (cash ?) 40% 36% 30% 22% 20% 6% 10% 5% 2% 2% 1% 0% 0% Payment PayPal Checks Credit Operator Private Prepaid Digital Others bank cards (N = 284) (N = 48) transfer billing label cards cards (N = currencies (N = 178) (N = 762) (N = 42) (N = 18) (N = 14) 6) (N = 3) Payment instrument De Nederlandsche Bank

Bank and non Bank and non Bank and non Bank and non- - - -bank instruments bank instruments bank instruments bank instruments 54% use only bank PI while 26% use bank and non bank PI (24% + 2%) 60% 54% 50% Percentage of respondents 40% 30% 24% 20% 12% 7% 10% 2% 1% 1% 0% Bank Non-bank Others Bank & Non- Bank & Non-bank & Bank & non- exclusively (N exclusively exclusively (N bank (N = others others (N = 4) bank & = 432) (N = 15) = 5) 188) (N = 59) others (N = 96) De Nederlandsche Bank

The case for financial privacy The case for financial privacy The case for financial privacy The case for financial privacy � PayPal and other PI are not always accepted by retailers (Amazon). But, a retailer who accepts PayPal always accepts a bank PI. � Why do consumers decide to use non-bank PI? � Convenience, security and person-to-person transfers. � We hypothesize that financial privacy is another motivation. � People use non-bank PI to protect their financial and personal data from banks, relatives, etc., for numerous reasons. De Nederlandsche Bank

Privacy and payment instruments Privacy and payment instruments Privacy and payment instruments Privacy and payment instruments Respondents who use non-bank PI care more about the exploitation of their personal data than the others. 90% 82% 80% Percentage of respondents 69% 70% 60% 51% 50% 43% 40% 30% 18% 20% 9% 10% 0% Cookies deleting (N = 605) Use of ad blockers (N = 382) Use of privacy enhancing web browser extensions (N = 110) Exclusive use of bank payment instruments (N = 432) Use of non-bank payment instruments (N = 203) De Nederlandsche Bank

Purchases, bank and non Purchases, bank and non Purchases, bank and non Purchases, bank and non- - -bank PI - bank PI bank PI bank PI The users of non-bank PI purchase more than the exclusive users of PI. 60% 54% Percentage of respondents 48% 50% 44% 40% 30% 30% 20% 11% 10% 6% 5% 2% 0% Less often than once More than once per Once per week Several times per per month month week Exclusive use of bank payment instruments (N = 432) Use of non-bank payment instruments (N = 203) De Nederlandsche Bank

Does financial privacy affect online Does financial privacy affect online Does financial privacy affect online Does financial privacy affect online purchases? purchases? purchases? purchases? � Dependent variable: frequency of online purchase: “less often than once per month”, …, to “several times per week”. � Ordered probit regression model. � Six groups of explanatory variables. 1. A binary variable: use of non-bank PI. 2. Convenience : financial info stored on e-commerce websites. 3. Privacy: ad-blockers, web browser extension, cookies deleting. 4. Risk: perceived risk of banking details being hacked or consulted. 5. Online activities: # PI used in purchases, # of passwords, etc. 6. Individual variables (sex, age, etc.). De Nederlandsche Bank

Results Results Results Results De Nederlandsche Bank

Results Results Results Results De Nederlandsche Bank

Does the use of non- Does the use of non -bank PI increase the bank PI increase the Does the use of non Does the use of non - - bank PI increase the bank PI increase the purchases of purchases of privacy purchases of purchases of privacy privacy- privacy -minded - - minded minded consumers? minded consumers? consumers? consumers? � Same regression but on two extreme populations: privacy-minded users versus others . � Privacy-minded users = privacy web browser extensions + ad-blockers + delete cookies � Others = they do not use extensions, ad-blockers and do not delete cookies. � Ordered probit regression model. � Six groups of explanatory variables. 1. A binary variable: use of a non-bank payment instrument (Y/N). 2. Convenience : financial info stored on e-commerce websites. 3. Risk: perceived risk of banking details being hacked or consulted. 4. Online activities: # instruments used in purchases, passwords, etc. 5. Individual variables (sex, age, etc.). De Nederlandsche Bank

Results Results Results Results De Nederlandsche Bank

Results Results Results Results De Nederlandsche Bank

Conclusion Conclusion Conclusion Conclusion � Consumers are more and more concerned with their financial data… � … and use in reaction more and more alternative payments systems (Bitcoin, etc.). � This paper is a first and very imperfect attempt to study and assess financial privacy. � Financial privacy should be further investigated by academics as it is a hot topic on the agenda of banks and public authorities. De Nederlandsche Bank