(Yicai Global) March 1 -- With the Chinese economy gradually recovering from the coronavirus epidemic, the digital economy that has performed well in the epidemic has attracted attention, and discussions on fintech and digital currencies are again heating up. How will the rise of fintech and big tech companies affect the financial eco-system? While stablecoins and central bank digital currencies (CBDCs) have received widespread attention, which design choices would lead to better outcomes, and what will be the effects on monetary policy and financial stability? 

Addressing those questions, China Financial Publishing House has published a new book titled Financial Infrastructure, Technological Innovation, and Policy Response: A Compilation of the Relevant Lectures of Zhou Xiaochuan. 

To promote further understanding of the issues discussed in the book by Zhou, the former governor of the People's Bank of China, Yicai Global had a conversation with Dr. Dong He, deputy director of the Monetary and Capital Markets Department of the International Monetary Fund. 

Yicai Global: How will the rise in fintech and bigtech companies affect the operating model and organizational structure of commercial banks?

Dong He: When we consider the potential impact from the rise of fintech and bigtech, there are two main considerations: economies of scale and economies of scope.

Economies of scale refers to the progressive reduction in average costs as the scale of operations increases, which motivates business entities to expand their size. Economies of scope refers to the progressive reduction in average costs with the shift from a single product to a diversified range of products, the underlying motivation being that the sharing of information in the financial industry can lower costs. Specifically, as long as the bank has grasped the basic information about a customer, it can sell many different types of products to this same customer and thereby increase profitability. For example, after a customer establishes a deposit account at a bank, the bank can offer the customer additional products and services, such as insurance and wealth management.

YG: How do bigtech and fintech companies affect banks and other financial institutions in terms of economies of scale and economies of scope?

Dong He: In terms of economies of scale, the advantages of commercial banks lie in relatively large corporate customers, while the advantages of technology companies lie in SMEs. The unit cost of commercial banks serving SMEs is high, and it is difficult to reduce the unit cost of such loans by expanding the scale. Technology companies can reduce unit costs by having a good grasp of data on SMEs through the use of digital technology. Therefore, the rise of technology companies should play a positive role in reducing the difficulty of financing for SMEs. In general, because commercial banks have powerful funding capacities, they still have a large advantage in terms of economies of scale.

The impact of fintech on the economies of scope of banks could be large because the advantage banks enjoy in regard to their knowledge of customers is no longer as great as it once was. For example, in the past, the control of customer information was an advantage the banking industry enjoyed exclusively. However, as advances are continually made in the digital age, social networking platforms and search engines have achieved a high level of grasp of customer information (digital footprints), so that this information is no longer exclusively enjoyed by banks. Indeed, social networking and e-commerce platforms may have better understanding of customer behavior than banks do.

In the future, it's possible that banks may no longer be the large, complex institutions with diversified product lines that they are today. Rather, they may gradually begin to collaborate with technology companies in different fields. For example, apps that sell insurance or wealth management products might appear on e-commerce or social networking platforms, with banks becoming part of the platforms, and loan and deposit functions might also only be individual apps on a social networking platform. This would represent a significant change from the current structure. Platforms could combine different financial services, and a wide variety of financial service providers can be expected to compete on the platforms.

YG: How do you judge whether competition from bigtech companies in financial services is fair? What's your view on cross-subsidization frequently used by platforms?

Dong He: In the digital age, it's typical for service providers to be organized as platforms, and such platforms are what economists call "two-sided markets." That is, a market in which the platform enables sellers and buyers to interact. These platforms bring together different communities of users seeking to interact with each other. In the case of WeChat Pay, the two groups of users are shoppers and merchants. We as shoppers would be interested in using WeChat Pay if most of the merchants from whom we buy products accept our payments through WeChat Pay. Conversely, the merchants would be interested in joining the WeChat network if most of their customers use WeChat Pay to make payments.

In these two-sided markets, we often see a high degree of market concentration. In China, Alipay and WeChat Pay dominate the retail mobile payments market. The first reason for market concentration is "externality": users benefit from the presence of those on the other side of the market. The second reason is "economies of scale." The more the users, the less is the unit cost of service. Because of network effects and economies of scale, we often end up with a situation of "winner takes all." To get there, platforms often grow thanks to the very low prices on one side of the market, which attract users on that side, and indirectly enables the platform to earn revenues on the other side. Thus as shoppers we pay nothing for using WeChat to make payments, but merchants are charged a merchant fee. 

Offering goods or services at a low price (or even for free) on one side of the market could be considered predatory pricing and naturally creates suspicion among competition authorities. In classical markets, it may be a strategy to put rivals out of business by weakening them financially, or simply by signaling the intention to be aggressive. Conversely, a very high price may suggest monopoly power. But in two-sided markets, such reasoning is probably no longer valid. In the digital age, market concentration per se may not be a sufficiently good indicator for lack of contestability. For example, service platforms usually encourage competition among merchants. New guidelines for competition policy as adapted to two-sided markets would require instead that the two sided of the market be considered together, as argued by the French economist Jean Tirole. Rigorous economic analysis is required on a case-by-case basis to judge whether there exists anti-competitive behavior.

In short, large-scale technology platforms not only enrich, but also complicate competition policies. Traditional indicators of market monopoly such as scale and market share may no longer be adequate or sufficient. Regulators and competition authorities need to formulate policies from a contestability perspective. How e-commerce or social platforms affect contestability, and whether large platforms can help improve consumer welfare and the productivity of companies on the platform, we need to study carefully and explore in depth.

YG: Will the entry of fintech and bigtech firms into financial services affect financial stability?

Dong He: To answer this question, we must first have a good understanding of the impact on financial stability of opening up the financial services industry, which is to say the relationship between increased competition and financial stability.

Generally speaking, competition in banking and financial services is good for society, provided that prudential regulation and supervision are adequate. Competition is a main source of economic efficiency and financial service is no exception. But we need to take into account how competition affects financial stability as finance is a confidence game and the financial system can be more fragile than other economic sectors.

Banking is a multi-product industry, and there are many frictions and barriers to entry that make competition in banking imperfect. In retail banking, switching costs, reputation, and branch networks act as entry barriers. In corporate banking, established relationships and asymmetric information are relevant frictions that explain why the market for small and medium-sized firms remains local. In wholesale and investment banking, competition is at the international level and market-based, so it may be fierce even if the market is concentrated.

Owing to behavior biases, increased competition in the financial system may lead to more fragility. For example, competitive markets tend to oversupply credit by relaxing lending standards and extending it to both good and bad risks, reinforced by consumers' tendency toward overborrowing, particularly when risk appetite is strong and the price of risk is low. Competition may also intensify risk taking by eroding the franchise value of incumbent institutions and diminishing incentives to monitor loans and maintain long-run relationships with borrowers. 

Thus, I would agree with the Spanish economist Xavier Vives that, it is plausible to expect a hump-shaped relationship between competition and stability with an intermediate level of competition being optimal for stability. Starting from a monopoly regime an increase in competition is beneficial because it spurs productive efficiency and innovation. Continued increase in competitive pressure may lead to a point where the benefits balance with higher fragility, and further increases beyond the point could be socially harmful. 

In this context, it is useful to note that the Chinese financial system is probably quite far from reaching the optimal level of competition, although the banking industry appears to have excessively homogenous competition in some respects and suffers from a lack of effective exit mechanisms. If the entry of bigtechs into financial services could promote healthy competition, that should be encouraged.

YG: With the opening of the financial sector and the entry of big tech companies into financial services, how should regulatory agencies incorporate increased competition into prudential policy design and implementation frameworks? How can competition policy and prudential policy be coordinated?

Dong He: The experience of Alipay and WeChat Pay in China shows that these two-sided markets for financial services can grow extremely fast, and the market structure could have fundamentally altered before the implications for financial stability are well understood. Therefore, the development of digital finance does indeed present challenges to regulatory policies.

Competition policy that eases entry, and increases contestability, should be accompanied by tougher prudential requirements. In a more competitive environment the solvency requirement has to be strengthened. This is because increased competition for funds aggravates coordination problems of investors and makes runs more likely, and it reduces the charter value of banks enticing them to take more risk. 

More broadly, the principle of "same functions, same risks, same regulation" should be adopted for the regulation of the financial service activities of fintech companies and big tech companies. A basic requirement is all financial service activities should be licensed. In practice, it can be difficult to determine what constitute "the same risk." Compared with traditional commercial banks, technology companies might be able to manage credit risks better due to enhanced identification and tracking capabilities, but they might have higher operational risks such as data leakage and network security. Apart from that, the question of how to regulate implicit leverage and liquidity risks of various financial products needs to be explored further. 

At this stage, the international regulatory community has emphasized the importance of monitoring and mitigating operational risks and cyber risks in the bigtech firms as they become an important part of the financial system. Over time, systemic importance and procyclicality could emerge from a number of sources, including from greater concentration in some market segments and if funding flows on the lending platforms were to become large and unstable. Authorities should regularly assess stability risks, bearing in mind the comparability of the functions performed, the level and types of risks involved, and the size of those activities. 

YG: The perspectives discussed above primarily involve microeconomic analysis. Viewed from the perspective of the macroeconomics of money and banking, will the situation be different?

Dong He: To answer this question, we must first understand how money is created in a modern economy. 

In a modern economy, money is created through the extension of credit. And this process defines the uniqueness of commercial banks. The two-tier structure of the modern monetary system refers to the relationship between commercial banks and the central bank. The liabilities of the central bank form the monetary base, which accounts for a relatively small proportion of the total money supply, but fulfills an important role as a fulcrum. Thus, we can visualize the monetary system as a triangular pyramid, with the central bank at the top of the pyramid and commercial banks in the middle of the pyramid, while the base of the pyramid is formed by non-bank financial institutions and other non-financial firms and households. 

Extending credit to create money makes the two sides of bank balance sheets closely interdependent and inseparable. The creation of liabilities is achieved by extending credit. Why is it that central banks do not extend credit directly to the general public and firms? This is actually a problem of efficiency. As a government-owned monopoly, central banks do not have a good grasp of the information of households and enterprises, therefore, they would enjoy no advantages in extending credit. By delegating the function to extend credit to commercial banks, overall efficiency is improved, because the distribution of commercial banks is relatively diffuse and commercial banks have a better understanding of households and firms. Deposits created in this way are actually an extension of central bank liabilities. Therefore, the relationship between the central bank and commercial banks is intrinsically close, with the latter being an extension of the money creation function of the former. 

Thus, the unique feature of commercial bank liabilities, and particularly demand deposits that can be used for payments, is the fact that they are redeemable on demand, transferrable to third parties, and denominated at par with central bank money. By comparison, it would not be credible for technology companies to issue transferable debt and commit to repay such liabilities that have a one-on-one conversion relationship with central bank money, i.e., to play a money creation function. In modern economies, the fiat currency is defined by central bank liabilities, and created primarily by commercial banks. Thus, it would be impossible for technology companies to replace the money creation function of banks unless they were also to become banks. In fact, some parts of the bigtechs have themselves become banks, such as the MYBank of Ant Financial and the WeBank of Tencent.

YG: How will e-money change the relationship between commercial banks and bigtechs?

Dong He: The rise of e-money has enabled bigtechs and other non-bank financial institutions to move to the forefront of payment services, and has allowed a separation of making payments, facilitating exchange, and defining the unit of account. With the rise of e-money and the popularization of fintech, third-party service providers can provide various payment services and become the interface with the household sector. However, these payment functions will continue to be settled through transfers between accounts held at commercial banks or the central bank, and these functions will not be replaced.

An illustration of the uniqueness of commercial banks is the diverging directions of "2C/2B" (to-customers vs. to-business) approaches of payment services. Payment giants like Alipay and Tenpay are currently working toward a transition to supply chain finance, i.e., an orientation toward 2B. But the transition has faced many difficulties. In contrast, commercial banks still enjoy very strong economies of scale in extending credit to firms. Since firms' cash flows are difficult to forecast, large payments typically happen irregularly and need to be facilitated by activation of credit lines, which banks can better accommodate than payment service providers. In other words, because commercial banks have the ability to create money, they have an intrinsic advantage in helping the corporate sector manage liquidity risks. Even in the digital age, there will continue to be an internal logic to the relationship between commercial banks and the corporate sector.

YG: Why has Facebook's plan to issue Libra triggered such a pronounced global reaction? Is there a future for non-sovereign digital currencies?

Dong He: From a technological perspective, the rise of digital currencies has been facilitated by breakthrough in the technology of tokenization. Tokenization and the use of smart contracts on secured networks allow for seamless automation of payments, securities trading and settlement, and other economic transactions. In such a tokenized financial system, physical distance and geographic space would lose meaning. Digital currencies could thus alleviate many of the frictions that underly the current international financial system, such as the correspondent banking relationships.

From an economic point of view, the rise of digital currencies reflects both demand and supply factors. On the demand side, digital currencies could be a solution to increase financial inclusion and to address existing pain-points in cross-border payments, such as low speed, high costs, and lack of transparency.  

On the supply side, breakthroughs in digital technology offer vast new opportunities for the introduction and use of custom-designed currencies. Digitization provides many opportunities for the design of new features of currencies adapted to the specific needs of the business introducing them. As more exchange of real goods and services takes place on large digital platforms that span multiple national economies (e.g. Alibaba, Amazon, Facebook, Google and Tencent), these digital sub-economies could take on the characteristics of optimal currency areas. Indeed, digital networks can customize the "acquirability," "transferability" and "redeemability" of the currencies in order to reinforce the business model of the issuers. 

Given the network effects, token use and exchange could be subject to large economies of scale, and the adoption of digital currencies can be potentially very rapid. Global stablecoins such as Libra therefore could potentially have large impact on monetary sovereign, financial stability, cross-border capital flows, and financial integrity. Regulatory challenges are daunting as global stablecoins involve financial services in multiple jurisdictions.

Nevertheless, there is a large degree of uncertainty as to whether independent digital currencies, those that have a different unit of account from fiat currencies, would have wide adoption beyond self-contained platforms. This problem is motivated by the question whether bank credit denominated in libra will be created and whether an eco-system of financial markets denominated in libra will be developed. Whether libra-denominated bank deposits can be created (libra "inside money") depends on whether is demand for liabilities denominated in libra. According to the "credit" theory of money, whether certain IOUs will be widely accepted (as transferrable debt or money) depend on trust, regulation, availability of Lender of Last Resort (LOLR), etc. In other words, the fiat currency is a "system of trust," including crucially the existence of a central bank. From this perspective, there is nothing in principle which prevents a digital currency from being created, but its wide adoption is difficult to envisage without the existence of a central bank, backed by the sovereign power.

More fundamentally, the future of private digital currencies vis-à-vis sovereign fiat currencies depends on the relative credit worthiness of big corporates or private digital networks vis-à-vis that of sovereign governments. Throughout history, private currencies emerged from time to time, but they tended to be plagued by financial instability and did not usually last long. As long as sovereign governments retain their creditworthiness, and central banks continue to run effective monetary policy, it is difficult to envisage that private digital currencies would gain wide domestic or international adoption. This does not rule out, however, that they could have limited acceptance or circulation in certain private e-commerce and social networks.

YG: Facebook's plan to issue Libra appears to have sparked central banks in various countries to accelerate research and development of central bank digital currencies. What are the primary challenges in the design and issuance of central bank digital currencies?

Dong He: In the face of rising challenges of private digital currencies, central banks need to move with the times and stay in the game of the digital economy. They should therefore continue to make their own liabilities (central bank money, or fiat currency) attractive to use by the private sector in the digital age. One feasible solution is to issue central bank digital currency (CBDC), which would allow the central bank to achieve the right balance in striving to protect privacy, respect legitimate anonymity, safeguard financial integrity, and maintain the effectiveness of monetary policy

In the digital age, it is conceivable that the payment system will feature many different types of payment instruments that satisfy different needs. CBDCs can be regarded as a "redundancy" or "backup" payment system — they complement other private payment instruments (e.g. e-money service providers such as Alipay and Tencent Pay) by providing a critical always-on fallback alternative in the digital age, just like the role cash plays in the analogue age. This is a critical public good that private service providers do not have sufficient incentives to provide. CBDCs also play the critical role of ensuring inter-operability of payments between different platforms, thus leveling the playground for different private digital networks. In other words, CBDCs are an important instrument in ensuring fair competition among digital platforms or networks.

The main challenges in designing and issuing CBDCs are operational: CBDCs should be extremely reliable, secure, and resilient against cyberattacks. They should always be available for use, particularly when there is no internet connection or when the telecommunications network is down. They should ensure a reasonable degree of anonymity and privacy for the users without compromising the ability by the authorities to enforce compliance with anti-money laundering and financing of terrorism laws and regulations. Central banks also need to carefully consider whether CBDCs should be accessible to foreigners and how the use of CBDCs cross-border can be made consistent with existing rule and regulations on foreign exchange restrictions and capital flow management measures.

These operational challenges imply that central banks are likely to be cautious when making decisions on the timing of the launch of CBDCs.

YG: What is your view of the concern that CBDCs will lead to "narrow banking" and negatively affect commercial banks? If central banks issue CBDCs directly to customers, will that lead to "narrow banking?"

Dong He: It is not clear why CBDCs would necessarily lead to narrow banking. Narrow banks can be a way or a channel to issue a CBDC, but they are not a necessary choice. 

Indeed concerns have often been expressed that CBDCs could negatively affect commercial banks ("disintermediation") in normal times and could make it easier to run on banks in stressful time. In my view, these concerns can be addressed by appropriate designs and institutional arrangements, for example, by making the CBDC a substitute of M0 or token-based digital cash. 

If the CBDC is designed as account-based, i.e. the central bank accepts deposits directly from the public including households and pays interest on such deposits, it may compete with commercial bank deposits and cause some disintermediation. 

It is also possible to envisage a narrow bank, which would have all the deposits it issues matched on the asset side with deposits at the central bank. The narrow bank would issue "narrow bank digital currency," or "synthetic CBDC." This is basically the same idea as stablecoins fully backed by central bank deposits.

However, in these two latter cases of deposit-based CBDC, it's possible to envisage that the negative effect on disintermediation can be mitigated by "tiering" of the CBDC deposit account. Under such a system, a relatively attractive interest rate is applied up to some quantitative ceiling, while a lower interest rate is applied for amounts beyond the threshold. This system allows assigning the payment function of money to tier one CBDC, and dis-incentivizing the store of value function through an unattractive remuneration rate on tier two CBDC.

Therefore, in my view, narrow banking could be a design choice for, but is not an inevitable outcome of, CBDC issuance. In any case, the central bank has instruments at its disposal to mitigate the potential negative effect of a deposit-based CBDC on commercial banks.

(The views expressed in this conversation are the personal views of Dr. He, and do not represent the views of IMF staff, management, or the institution.)