By Tommaso Mancini-Griffoli
The IMF recently published a paper on the potential effects of fintech on the financial sector and how regulation should adapt: “Fintech and Financial Services: Initial Considerations,” co-authored by Dong He, Ross Leckow, Vikram Haksar, Tommaso Mancini-Griffoli, Nigel Jenkinson, Mikari Kashima, Tanai Khiaonarong, Céline Rochon, and Hervé Tourpe
Question 1: The term “fintech” sounds very general. What does it cover?
Fintech—the application of technology to the financial sector—is a broad term. Artificial intelligence, big data, biometrics, and distributed ledger technology, such as blockchains, are just a few of the technologies that fall under the fintech umbrella.
Financial services are already beginning to change as a result of fintech. Peer-to-peer funding is making some headway, messaging applications are offering payment services, cybersecurity risks are increasingly analyzed by machines that track online behavior, and credit scoring is gradually reflecting the vast amount of data on borrowers. Financial activity may shift away from traditional intermediaries like banks toward networks in which market participants do business directly with each other.
End-users may benefit from these changes, but the question for policymakers is the impact fintech could have on the structure and role of banks, financial stability, customer safety, financial inclusion, and monetary policy transmission.
Question 2: Can you outline some big trends in fintech?
There are two main forces that could be especially disruptive.
The first is the explosion of available customer data—on financial transactions, constraints, and goals but also on identity, preferences, habits, and connections. There is a huge opportunity for change depending on who can aggregate and control this data and who can link it to easy-to-use interfaces and applications that may already be in use.
The other big change relates to payments. Ever since the invention of double-entry bookkeeping, we have perfected and increasingly relied on the account-based system of payments. That is, instead of handing cash to each other, we have instructed banks to alter entries in their ledgers. This requires costly identity verification, accounts, liquidity and risk management, and payment clearing and settlement services ultimately provided by central banks. Currently, cash is nearly a rarity.
But fintech could change all that. Depending on the course of technology, and of central banks, we may start to use the electronic equivalent to cash—electronic tokens—even for payments across borders. Imagine the potential changes to the banking system, to the role of trust, to the speed and cost of transactions, even to the expansion of the decentralized service economy requiring small-value payments within and across borders. The implications could be huge.
Question 3: So what could be the effects beyond new and better services?
This IMF paper introduces a framework for thinking about the effects of fintech on market structure. The focus is first on identifying shortcomings of current services—areas in which better services thanks to new technologies could cause a jump in demand. The focus then shifts to how these same technologies could affect the need for financial intermediaries, as opposed to standard market platforms; the organizational structure of these intermediaries; and barriers to new entrants. The hope is that the framework will still be useful in future work.
Intermediaries—such as banks, firms specialized in messaging services, and correspondent banks clearing and settling transactions across borders—are common in financial services, but will face significant competition. Some may partner, others will change, and a few will bow out.
New technologies such as identity and account verification could lower transaction costs and offer more information on counterparties, making intermediaries less relevant.
Existing intermediaries may be pushed to specialize and outsource well-defined tasks—possibly including customer due diligence—to technology companies.
And intermediaries are likely to change the way they are organized. The degree to which they maintain control over their clients’ data will determine how much they can cross-sell other financial products. But the value of access to other types of data as well, such as on preferences and habits, may favor partnerships and bundling deals with other types of companies, such as social media firms.
Barriers to entry will also evolve. On the one hand, the lower cost of offering financial services, such as by automating manual back-office tasks, is likely to favor entry. On the other hand, trust will still be vital, if not in financial intermediaries then in financial services—how stable and secure they are. Such trust will still require a large investment on the part of service providers, which will bar entry for some. In addition, network effects will remain relevant: new firms’ ability to communicate and transact on existing financial networks or markets will determine how easily they get up and running.
Question 4: What are the greatest challenges facing regulators and how might they respond?
Regulation is intended to protect consumers, prevent monopolistic behavior by banks, and promote financial stability and integrity. Fintech introduces risks on all these fronts.
Take algorithms—machine learning—for instance. If you rely on them for trading, you could be exposed to the risk that they will become procyclical by trading all at the same time; that they will fail; or that they will be compromised in a cyberattack. Will regulators have to be software engineers who can check code?
Another challenge is increasing access to customer information. More readily available data clearly has benefits, but the flip side is a greater risk of compromised digital identities. If someone gains access to your retina scan, which you use to secure transactions, you cannot change your retina as you can a compromised credit card number.
New rules may be needed to ensure sufficient consumer safeguards, including privacy protection, and to guard against money laundering and terrorism financing, particularly when it comes to decentralized virtual currencies.
Other challenges include the shift from regulating entities to regulating activities. As financial activities shift to networks, they cannot be effectively regulated solely by focusing on intermediaries that no longer provide them. Striking an appropriate balance between entity- and activity-based regulation will be a key challenge for regulators.
International cooperation will be critical. Advances in technology know no borders, and such cooperation will ensure a level playing field and effective cross-border regulation.
Finally, regulation should continue to function as an essential safeguard to build trust in the stability and security of the networks and algorithms.
Question 5: What are some tangible predictions in the new paper?
A key part of the paper focuses on cross-border payments. The idea was to pick a slice of the financial sector, close to the international mandate of the IMF, to consider the impact of fintech through the lens of the paper’s analytical framework.
People are generally unhappy with cross-border payments. Sending money abroad is slow, unpredictable, and costly. These shortcomings in part reflect the limitations of technology. Without an international central bank, most payments are cleared and settled by private correspondent banks.
Fintech solutions come in various flavors. Distributed ledger technology and artificial intelligence could lower back-office costs, help track payments and reconcile invoices, and manage liquidity. It could also lower the cost of complying with regulations through easier capture, checking, and sharing of customer identity.
But the larger impact on market structure and regulation would come from the use of tokens for transactions across borders. These networks could bypass the large commercial banks with the press of a button and eliminate the need for separate communication networks.
Such networks may never take off for a number of reasons, but the implications for market structure are significant if they do. Just as email eliminated the need for ways to send letters domestically and internationally, payments could change dramatically if we start using tokens.
Question 6: What work on fintech has the IMF produced to date?
This is the second Staff Discussion Note on fintech published by the IMF. The first came out in January 2016 on the topic of virtual currencies, such as bitcoin, and whether they would take off and revolutionize the world. The short answer is no.
This paper takes a bit of a step back and proposes a framework for thinking about how fintech could affect the financial sector and how regulation might need to respond. It then applies that framework to gauging implications for cross-border payments—a subject at the heart of the IMF mandate.
Question 7: Clearly there are a lot of emerging issues that the industry is still grappling with. With that in mind, how do you see the IMF’s work on fintech evolving?
There is potentially an enormous amount of work to do in thinking through various scenarios of fintech development and adoption.
Fintech could affect any of the five key needs for financial services, as seen from the perspective of end-users: paying, saving, borrowing, managing risks, and getting advice on financial services. In this Staff Discussion Note, we tackled only the subset of payments.
Future work will likely focus on areas with the potential for a significant impact on financial stability and inclusion and monetary policy transmission.
Fintech’s effect on monetary policy transmission has generated an interesting debate, summarized in the paper, on whether central banks should introduce digital currencies. Would monetary policy transmission change? Would banks disappear? Could privately issued virtual currencies compete or possibly even do better than traditional currencies? Would weak currencies—issued by central banks lacking credibility—disappear? Would a first mover’s currency become the new reserve currency? Who said economics was boring?!