My name is Stuart Rubinstein and I am President of Fidelity Wealth Technologies and Head of Data Aggregation. In this role, I oversee the team focused on helping Fidelity and other institutions enable consumers to securely share account data and documents with third parties. Fidelity is a leading provider of investment management, retirement planning, portfolio guidance, brokerage, benefits outsourcing, and other financial products and services to more than 30 million individuals, institutions, and financial intermediaries with more than $7 trillion in assets under administration. Our goal is to make financial expertise broadly accessible and effective in helping people live the lives they want.
I will focus my testimony for this hearing on an issue I first worked on over 20 years ago: financial data aggregation services and ways we can make data sharing safer and more secure.
Fidelity has a unique perspective on financial data aggregation practices and necessary protections for customers. We are on all sides of this issue: we are an aggregator of data for third parties, we are a significant source of data for aggregators acting on behalf of our mutual customers, and we offer a data aggregation service for our retail customers and retirement plan participants. This perspective gives us a thorough understanding of the benefits of financial data aggregation, but also of the very real cybersecurity and privacy risks that current data aggregation industry practices create.
Financial data aggregation in this context refers to services that, with customers' consent, collect financial information from their various bank, brokerage, and retirement accounts, along with other sources, to be displayed and processed in an aggregated view. An example of this kind of service might be a budgeting and planning smartphone app. Consumers use third party applications that leverage data aggregation because they value tools to help manage financial planning, budgeting, tax preparation, and other services. As part of our focus on helping our customers, Fidelity works to make it possible for customers to access the services they want to use—including third party aggregation-based services. To that end, customers have been able to use their Fidelity data in third party applications for many years. However, the cybersecurity environment has significantly changed over that time and we have a responsibility to protect the very sensitive personal financial data and assets of our more than 30 million customers from misuse, theft, and fraud.
Current data aggregation practices make this challenging, because they rely on consumers providing their financial institution log-in credentials (i.e., username and password) to third parties. Those third parties, typically data aggregators, then almost always employ a practice known as "screen scraping." At its most basic, screen scraping involves the use of computerized "bots" to log-in to financial institution websites, mobile apps, or other applications as if they were the consumer. Once the bots have access to the site or app, they "scrape" customer data from the various screens to be presented on a consolidated basis, along with information scraped and collected from other sources.
There are two consumer data security problems with this practice. First, as a matter of basic security consumers should not be asked or required to share their private log-in credentials in order to access a third party service. Doing so creates cybersecurity, identity theft, and data security risks for the consumer and financial institutions. Unfortunately, we know that due to years of this practice, financial institution log-in credentials are now held by a myriad of companies. Some are likely very secure, while others may not be secure at all. Given this, allowing third parties to log-in using these credentials as if they are the customer creates significant risk of cyber-fraud. Because consumers go directly to data aggregators or their commercial clients and not their financial institution, the financial institutions never really know if the activity has in fact been authorized by the customers or if the customer credential has been compromised and a criminal is using the data aggregation service to test the credential's validity and illicitly gather data.
Second, screen scraping may result in access to data fields far beyond the scope of the service a third party offers the consumer—including personally identifiable information (PII) about consumers and in some cases their dependents. This means third parties have access to fields of information often used by financial institution call centers to identify customers. For example, if a consumer provides his or her log-in credentials to a budgeting app, that app potentially has access to sensitive personal information like customer dates of birth and dependent names and dates of birth, all of which might be data financial institutions use to verify customer identities online or over the phone. Collection of information beyond what is needed for the service the consumer has elected creates unnecessary risk. And all of this adds up to an array of risks financial institutions must navigate to protect the integrity of their systems and the assets of their customers.
In considering the challenges described above, Fidelity developed the following five principles that we believe should guide industry in creating better data sharing solutions:
We believe that embracing these principles will better protect consumers, aggregators, and financial institutions, and facilitate more efficient data sharing practices.
Fortunately, although the risks and challenges of the current system are serious, there are steps financial institutions and aggregators can take together to improve the data sharing ecosystem. The financial services industry is employing technological solutions for the secure exchange and access of financial information. These technologies involve the implementation and use of application programming interfaces ("APIs"), which are provided by the financial institution to aggregators and other third parties. An API works in conjunction with an authentication process that is handled by the financial institution. There are authentication processes, for example "open authorization" ("OAuth 2.0"), that do not involve sharing of account access credentials with third parties. Consumers who want their data aggregated sign into their accounts at the financial institution's website and provide authorization for third party aggregators to access their financial data. The financial institution and the data aggregator then manage that connection through secure, encrypted tokens that are provisioned for the specific connection.
There are several compelling consumer and data security benefits for moving to APIs. First, it keeps log-in credentials private and secure by eliminating the need for consumers to share log-in credentials with third parties. This reduces the cyber, identity, and personal data security risks that exist when a consumer shares private log-in details with a third-party. Second, it puts the consumer in the driver's seat by giving consumers greater transparency and control of their data 5 by allowing consumers to provide unequivocal consent and instruction to share their data with third parties. Third, it allows financial institutions and aggregators to agree on what data should be shared and avoid over-scraping. Fourth, it eliminates the need to reconfigure aggregators' systems every time a consumer changes his or her username or password or the financial institution updates its webpage. Fifth, it removes the traffic-intensive screen scraping activity from financial institutions' web sites and other digital properties, returning that capacity to the individual consumers for whom those sites were created. Finally, it enables the consumer to monitor the ongoing access and instruct their financial institution to revoke the consent if desired.
In November 2017, Fidelity announced its own API solution for data sharing called Fidelity AccessSM. Fidelity Access will allow Fidelity customers to provide third parties access to customer data through a secure connection without providing log-in credentials. Fidelity Access will include a control center, where customers can grant, monitor, and revoke account access at any time. We have been working closely with aggregators and other third parties on adoption of this solution.
Of particular note, eMoney Advisor, Fidelity's affiliate that offers its own aggregation service, is committed to working with other financial institutions that offer APIs. By championing the exclusive use of APIs to facilitate customers providing third parties access to their financial data, we hope to show leadership by taking action to better secure our customers' data.
In addition to our own efforts to address the problems with data aggregation, we have been working with a wide array of industry and public sector stakeholders. We support many of the data sharing and aggregation principles that have been put forth:
These efforts to provide guidance have brought many of the challenges and risks associated with data aggregation to the fore and encouraged healthy debate on how to solve them.
Despite the general consensus that the status quo is untenable and the industry should move to safer data sharing technologies, there are roadblocks that prevent wider adoption of APIs and other solutions. Here are what we see as the most challenging:
Until all industry participants—aggregators, FinTech firms, and financial institutions—are prepared to overcome these challenges in a responsible manner, we will not move as swiftly as we otherwise could to adopt safer data sharing technologies.
In July 2018, the U.S. Department of Treasury issued a report entitled "A Financial System that Creates Economic Opportunities: Nonbank Financials, Fintech, and Innovation" (hereinafter "Treasury report"), in which it discussed at length the public policy challenges facing the data aggregation industry. We agree with much of the Department's analysis, including its focus on security, consumer consent and monitoring, revocability of consent, and liability. In particular, we would like to share views on the following data aggregation recommendations in the report:
In order to reduce the complexity of complying with 50 unique state data breach notification laws, the Treasury Report recommends that Congress enact a federal data breach notification law that would preempt state data breach laws.
Fidelity View: To simplify the increasingly complex 50-state data breach notification regime, Fidelity would support a federal data breach standard that preempted state data breach notification laws and included robust consumer protections. As part of the legislative process, Congress and any relevant regulatory agency should ensure that companies have adequate time to investigate a potential breach and that consumers benefit from a required and timely notification related to a breach of their personal information.
Thank you again for the opportunity to testify and I look forward to answering your questions.