FinTech and Consumer Decision-Making in the Information Age Bruce Carlin∗ Arna Olafssony Michaela Pagelz November 2020 Abstract We exploit the release of a mobile application for a financial aggregation plat- form to analyze how Financial Technology (FinTech) adoption changes consumer financial decision making. Our sample consists of individuals that had been using the platform via a desktop computer long before the mobile app was released. The app reduced the cost of accessing personal financial information, and this was re- sponsible for a significant drop in the use of expensive consumer credit and late payment fees. The leading explanation for our results appears to be mistake avoid- ance, which is supported by a significant reduction in non-sufficient funds (NSF) charges after the app was released. JEL classifications: G5, D14, D83, G02. ∗Department of Finance, Jones School of Business, Rice University, Houston, TX, USA, & NBER. [email protected] yDepartment of Finance, Copenhagen Business School, 2000 Frederiksberg, Denmark, the Danish Finance Institute, & CEPR. ao.fi@cbs.dk zDivision of Economics and Finance, Columbia Business School, NY, USA, NBER, & CEPR. [email protected] We thank numerous seminar participants, and our discussants and conference participants at the AF- FECT Conference University of Miami, University of Kentucky Finance Conference, Santiago Finance Conference, Cerge-Ei Prague, 6th ITAM Finance Conference, and the AEA. This project has received funding from Danish Council for Independent Research, under grant agreement no 6165-00020. This project has benefitted from funding from the Carlsberg Foundation. We are indebted to Meniga and their data analysts for providing and helping with the data. We also thank Fedra De Angelis Effrem and Andrea Marogg for outstanding research assistance. 1 Introduction Does the availability and adoption of new financial technology equip consumers to make better financial decisions? Ostensibly, if people have easier access to information, they should be able to avoid mistakes (Stango and Zinman, 2009; Jørring, 2019). However, while the rate of technology adoption is straightforward to quantify (Carlin et al., 2019; Anderson, 2015), measuring its economic impact is challenging. It is difficult to find settings, especially natural experiments, in which it is possible to calibrate how technology affects people’s behavior and their outcomes. FinTech has advanced at a rapid rate in retail markets and is receiving growing atten- tion (Goldstein et al., 2019). Personal financial management platforms (Gelman et al., 2014; Baker, 2018; Olafsson and Pagel, 2018; Kuchler and Pagel, 2018), mobile payment systems (Agarwal et al., 2019), and robo-advisors (D’Acunto et al., 2019; Loos et al., 2019) are just a few examples of the rapidly changing landscape for consumers. However, while FinTech advances have been shown to be valuable to their innovators (Chen et al., 2019), it remains an open question how consumers are affected. In this paper, we investigate how access to mobile financial apps affect consumer financial decision making. We use individual transaction-level data from a financial ag- gregation platform in Iceland called Meniga. Meniga allows users to link all of their checking, savings, and credit card accounts, and view all spending, income transactions, and account balances in one place. A considerable fraction of adults in Iceland use this service and the user population appears to be representative of the overall population (Olafsson and Pagel, 2018; Carvalho et al., 2019). As we describe in the paper, using data from Iceland has several advantages for our purposes here. We exploit an exogenous shock that made accessing the online platform easier. Before November 2014, access to the personal financial management software was possible only via the Internet on a desktop or laptop computer. However, on November 14, 2014, a mobile application was released, which gave users easier and remote access to bank 1 account information. The effect of the mobile app release on the propensity to log in to the platform is shown in Figure1. Each dot represents the raw average fraction of individuals who logged in using any method each month. There is a clear discontinuous jump in the propensity for users to log in to the platform around the mobile app release. {Figure1 around here} Our data set contains time-series information about the frequency and method of access to bank information (computer vs. mobile app), demographics, expenditures by category, income, use of consumer credit (credit cards and checking account overdrafts), and the resultant financial outcomes (consumer debt and bank fees).1 We perform an event study to compare each consumer’s financial behavior before and after they start using the mobile app. In our primary specification, our estimates represent the average monthly effect on bank fees during the first year after the app was released.2 To avoid selection bias, we only analyzed individuals who used the platform for at least a year prior to the app release (37 months on average).3 We include individual fixed effects to control for all time-invariant individual characteristics and calendar fixed effects to take care of seasonal variation. We provide evidence that no other confounding events took place around the same time. After the mobile application was introduced, consumers paid fewer bank fees. Plots of the raw data show that monthly late fees, overdraft interest, and total fees decreased after the app was introduced (Figures2-4). It is likely that the release of an app lowered the cost of logging in to the platform and thereby changed people’s behavior. Indeed, 1Note that, in Iceland, checking account overdrafts are the way most consumers roll over unsecured high-interest consumer debt. Individuals pay interest on overdrafts, but there is no discrete overdraft fee. Credit cards are common but are typically repaid in full each month rather than used to roll over debt. 2We focus on a short period (one year) after the release of the mobile application. We also consider other lengths of time after the app release (3 months, 6 months, 18 months, and 2 years) for the same group and show that the results are qualitatively robust. 3We exclude individuals that were not already using the platform twelve months prior to the mobile app release to address the concern that some users may have started using the platform because they expected an app to be released soon. 2 based on the raw data there appears to be a monotonic, decreasing relationship between the number of logins and the fees that consumers pay (Figure5). 4 Regression estimates suggest that the release of the app reduced average monthly bank fees by $2.22 and average late fees by $1.38. Given that individuals paid $21.95 in average monthly bank fees and monthly late fees of $7.16 before the app was released, the new technology was associated with a 10.1% reduction in monthly bank fees and a 19.3% decrease in late fees. In our sample, the average consumer rolled over $487 in overdraft debt each month. Based on our results, the twelve-month average cumulative decrease in overdraft interest was $50, which represents a 10.27% decrease. As such, our results are statistically significant and economically compelling. {Figures2,3,4, and5 around here} Previous work finds support that incurring late fees and paying overdraft interest are often associated with financial mistakes (Stango and Zinman, 2009; Jørring, 2019). To make that conlusion, the authors observe each consumer’s balance at the time the fees are incurred and confirm whether they were avoidable. Unfortunately, in our dataset, infor- mation about fees and interest payments is only available at a monthly frequency, which precludes a similar exercise. So, to investigate whether the app introduction was associ- ated with better financial health and fewer mistakes, we analyze the non-sufficient funds (NSF) charges that consumers paid. As discussed by Carvalho et al.(2019), accruing NSF fees are due to financial mistakes. In our sample, we find that the release of the mobile application was associated with a 14.3% drop in the number of NSF fees, which provides evidence that the new technology helped consumers avoid making financial mistakes. The Meniga platform is exclusively for informational purposes and does not allow for any financial actions (e.g., paying bills). As such, release of the mobile application likely made access to information less costly and promoted more frequent information 4See Olafsson and Pagel(2017) for an in-depth analysis of the within-individual patterns of logins and personal finances. 3 acquisition. It also may have made financial information more salient.5 However, the software did not provide nudge-based interventions (Stango and Zinman, 2014; Karlan et al., 2016; Medina, 2018).6 In the presence of time costs, a reduction in the costs of paying attention to finances may result in a reduction in consumer debt. That being said, the fact that consumers reduced their debt after getting easier access to financial information may contribute to our understanding of why high-interest consumer debt exists in magnitudes that are puzzling from the perspective of standard preferences in life-cycle consumption models (Angeletos et al., 2001; Carroll, 2001; Laibson et al., 2003; Zinman, 2015). Our results may speak to non-standard preferences and overconsumption problems as explanations for the initial use of consumer debt (Laibson et al., 2007), rather than consumption smoothing in response to permanent income shocks when funds are tied up in illiquid savings (Kaplan and Violante, 2014) or consumption smoothing in response to transitory income shocks (Keys, 2010; Sullivan, 2008). The remainder of the paper is organized as follows. In Section 2, we describe the data and provide summary statistics. In Section 3, we explain our identification approach, re- port our main results, and discuss their robustness. Finally, Section 4 provides concluding remarks.