Nov 23, 2018 08:45 PM

Ex-Central Bank Head Warns About Tech’s Influence on Finance

Zhou Xiaochuan, former governor of the People's Bank of China. Photo: Caixin
Zhou Xiaochuan, former governor of the People's Bank of China. Photo: Caixin

Editor’s note: The development of financial technologies such as digital currency and electronic payments has been spearheaded mainly by China’s privately owned technology giants. The rapid rise of these companies in the financial industry has posed an unprecedented challenge to the traditional financial industry, as well as to regulators, who have struggled to calibrate their oversight of these new developments. For example, Ant Financial, the online payment affiliate of Jack Ma’s Alibaba Group Holdings Ltd., serves more customers than some of China’s largest banks. It also operates the world’s largest money market fund, which has nearly 2 trillion yuan ($288.5 billion) in total assets under management.

Chinese regulators have become increasingly wary about the expansion of private fintech firms due to concerns over stability, security and user privacy.

In a speech given on Sunday at the 9th Caixin Summit in Beijing, Zhou Xiaochuan, the longest-serving governor of the People’s Bank of China (PBOC), suggested taking a cautious approach toward new technologies such as blockchain. The former PBOC governor, who retired in March after 15 years in the position, also blasted some fintech companies for focusing too much on financial gain rather than improving services. He expressed concerns that the “winner-take-all” style of expansion by leading internet firms may undermine competition and hinder innovation. He also called for private companies to give greater consideration to the public good when building financial infrastructure.

What follows is an English translation of an excerpt of his speech.

The development of digital currencies and e-payment services is moving forward very quickly, and everyone expects that in the future, e-payments will change the payment industry to a great extent.

First, I would like to remind you of the relationship between finance and information technology (IT). In the past, some studies have said that finance companies, especially commercial banks and insurance companies, are in principle IT companies as they mostly deal with data, and currency is already mostly digitized anyway.

In the case of China, our cash is only 5-6% of the entire currency stock, the rest of our money is stored in computers in the form of zeroes and ones. When you give a company a loan, you draw on historical data of a certain type of company in a certain location and certain industry to determine whether or not to make the loan and what risk premium to assign. That is all data processing. The insurance industry is even more so.

But any IT company needs a user interface, that is to say, eventually they will all interact with users in one way or another, and we can see this user interface in retail outlets or service counters.

Of course, I personally think that this perception goes a bit too far. But I do think that finance is in fact doing similar things to the IT industry, and finance firms can be seen as quasi-tech firms. Traditional finance companies have always been major users of technology and fintech, and are the major users of IT and fintech now. We can spend time looking at the evolution of IT in finance chronologically, and we can see that finance companies have been the largest buyers and users of IT technology.

We can look at this from the four following perspectives:

The first is computational capabilities — from early computers to cloud computing.

The second is storage capabilities. In the past, the finance industry was the largest purchaser of storage devices. But now, we have more audio, video and speech data, occupying more storage space than structuralized numbers and text. The finance industry is not the largest user anymore, but still an important user.

Then there is network access. The finance industry is heavily dependent on networks. In the early days we didn’t call it “network,” we called it “long-distance communication.” Now everyone knows that it is a network as well, it was just a different generation. The finance industry is also pretty much the largest user of data, but of course now it’s not called “database users” but “big data.” In summary, we can see that from a chronological perspective that finance and IT have had a close relationship and have promoted the development of the other.

Based on this, we can say finance is benefiting from the development of the IT sector, as technology has enhanced the quantity, quality and efficiency of financial services, and so the finance industry should welcome competition and new technology. Of course, financial institutions will try to protect themselves and compete against each other, so sometimes a company could express its opinion or objection toward certain technologies. But in general, we must look at the overall direction and the relationships between finance and IT clearly.

Secondly, I would like to talk about supply and demand. Sometimes fintech and IT have different angles on supply and demand. We think that one key field is payments, which is a pillar of the national economy.

Looking at the payments field, from the point of view of the payment system, what we really need, first, is efficiency. Second, the costs must be kept low. We cannot say that the cost of doing one payment is too high. Third is safety and stability. There cannot be any problems, in both the protection of privacy and transaction reliability. Should the safety or stability of payment systems encounter a problem, it would have a major impact on the economy and society.

Of course, new technology appears all the time. After someone invents a new technology, he wants to sell his new technology. He might sell it from the angle of promoting the characteristics of the technology or from how it is useful, saying “this technology has these characteristics that are useful for you.” But these two aspects are not always the same. We say that blockchain technology developed into distributed ledger technology. This technology has characteristics that might be useful in the financial field, in financial markets, in financial transactions and in the future may have other prospects. However, it’s also important to see that while some of these characteristics are important for the future of the payment system, there are some that are not necessarily key features for the demand side.

For example, whether decentralization is a core problem of the financial and payment systems should still be debated and studied.

Regarding the development of technology, we can say that part of technology is developing in steps. It can help us move up a step, regardless of how high or low the step might be, some might be 15 centimeters (5.9 inches), some might be 20 centimeters, some are half a meter, and the gap might be huge. But there are also disruptive developments and breakthroughs that we must pay close attention to that fundamentally change the way traditional businesses operate. So, we have to make judgements and be able to tell the disruptive ones apart from the gradual ones. Pay attention to this.

Because the supply side and demand side sometimes have different views, there must be some intermediation and coordination between the two. During this process, we will see some distortions and risks. I will focus on three aspects:

One is the use of new products and new technologies as the main tools for speculation and making money. Someone might start to trade the technology on the market very early, thinking that trading it on the markets would yield a big profit. This could distort the market and even cause more damage.

Another trend is that some companies did not focus on the potential of the technology for financial services, but instead focused on how to get more money, especially in the pockets of consumers, the money in deposit accounts, and focus too much on considering if it could absorb deposits like commercial banks do. We started issuing third-party payment licenses around six years ago, but among the roughly 200 companies we issued licenses to, many did not truly posses an interest in developing payment technology, increasing efficiency and reducing costs. Instead, they cared only about getting the prepayment money, as if it was its only purpose. This could create a lot of distortion. Recent failures in peer-to-peer (P2P) companies’ fund pools are a similar kind of distortion.

The third point is one that Caixin is leading research on: the “winner-take-all” phenomenon in IT and internet companies. This is not what we originally hoped for. We hope that through competition, the best technology can stand out and become widely used. That is why this phenomenon is a problem.

Thirdly, I want to talk about the distinction between digital currency and online payments. There is much debate on this topic, and it is often hard to come up with a definition. The Bank for International Settlements (BIS) issued a report on central banks and cryptocurrency that started by discussing the problem of terminology. We are right in saying that terminology is not a simple issue of technology, because it can affect how we look at a problem, which angles we use to view the development of these technologies.

Of course, the first, obvious angle is whether the currency created is physical or digital. To be honest, I have just mentioned that most currencies are already digital. But some would argue that “digital currency” can only refer to cryptocurrency with blockchain technology, and those that are not based on the blockchain cannot be described as a digital currency. I would say this is still a matter of debate, there isn’t an agreement yet.

The second distinction is about whether digital currency is based on tokens, we say “token-based” or “account-based.” In the case of China’s development, from credit cards in the past and now mobile payments and QR codes, they are all account-based. So that is another choice to make.

Another distinction to make is whether the payment tools and digital currency in question are designed to support retail or wholesale. Choosing wholesale means that this system will also provide central banking functions. Some of the retail-oriented digital payment systems are in fact location-specific, for example, college campus payment systems. This difference is due to the division of labor between the central bank and commercial bank, and the third-party payment systems. It also involves the consideration of system security, stability and reliability.

After BIS made these distinctions, they classified digital currencies into central bank digital currencies and private sector digital currencies. Of course, BIS’s definition on private sector is broader than ours and includes the commercial banks, regardless of whether they are state-owned or not. They belong to the private sector as long as they are not central bank-related. Of course, we can also have public-private-partnerships (PPP) in the digital currency field.

In short, because the mission of central banks is to maintain a stable currency value and a stable financial system, it may pay attention to this aspect. If there is no reasonable mechanism, regulations and incentives for the private sector, it may only care about market share, efficiency and cost, but not necessarily about currency stability.

Therefore, after a period of exploration, if the value of the currency is not stable, then some people have come up with a digital currency that is pegged to the central bank currency, called “stable coins,” so it also shows that such an experience will produce such a demand.

I think BIS has an incomplete set of terminology and categories and needs to consider other specific features. The categories can be large or small. One of the things BIS did not cover is credit-based or debit-based. We see that many of the mainstream third-party payment services in China are still debit-based, but many P2P companies are also making loans to support mainstream debit payments. This is another difference.

I have already talked about another distinction, that is whether the coin is pegged to another currency or not. Recently, people have been watching the pegged digital currencies closely. Another distinction is whether it is encrypted. Actually, if you look close at digital payment and digital currency, pretty much all of them are encrypted to a degree. But the encryption is in different places. Some would encrypt its owner; some would encrypt the transaction information; some would encrypt the token transactions. In a sense, you cannot avoid encryption completely or it would be too easy to hack.

Some people say, in theory, some of the cryptocurrencies can be considered completely safe. But complete safety is not achievable: as you upgrade your defense, others upgrade their attack; everyone is susceptible to attack.

We should also mention data retention. To what extent should we allow past data to be saved? Data retention would involve how to settle disputes, but more so would involve how to protect privacy.

After we are clear on these notions, we could see that there might be several paths ahead for digital currency and e-payments and they will move forward in competition. The future can be full of uncertainty; this has posed a great challenge to the central bank and regulators. I don’t work at the People’s Bank of China anymore, but I know a digital currency research institution was started three to four years ago, when I was there. It was responsible for fintech research and digital currency research. The PBOC is capable of conducting research in this field, but you can never be sure that your research will always yield the optimal solution. Technology is always evolving; it can be risky to choose a certain technological path.

One way to cope with this is to design a system with multiple competing developers. But it is also important to set up a standard procedure to make sure the developers produce controllable results. It is possible that a plan from the developers will do large-scale damage to the economy and to society. So we have to make sure they are controllable.

A famous case is the “regulatory sandbox” approach advocated by the Bank of England. But the sandbox might not be enough for making large-scale technology decisions; it is better for smaller experiments and a selection of technologies.

We should also realize that some services should be classified as “financial infrastructure,” which is, by nature, a public good. In addition to stability, it also requires a high level of security. That is because if infrastructure fails, it has a serious impact.

For example, printing notes, minting coins, and clearance systems are all obvious parts of financial infrastructure. A social credit system, if there is going to be one, would be part of the financial infrastructure as well.

The Financial Stability Forum and BIS, created by the G-20 both have sub-commissions on financial infrastructure. The Committee on Payments and Market Infrastructures (CPMI) is a commission that focuses on payment systems, market infrastructure and financial market infrastructure. We have to realize financial infrastructure share similar characteristics with other infrastructure in the economy.

If it has a public nature, does a public institution have to shoulder responsibility? I personally do not think so. The private sector can provide infrastructure sometimes. Of course they have to do so under the guidance and regulation of the government; the PPP model is also possible.

I will stress one thing: after undergoing a period of training, if the private sector wants to participate in providing financial infrastructure, it must have “public spirit.” “Public spirit” refers to the willingness to serve the public interest, which means the company will not use the privilege and advantage gained from the infrastructure it provides to seek profit. For example, if a company sells the user data derived from its infrastructure for profit, this is a lack of “public spirit”. Companies that are proven to have “public spirit,” or are in the process of cultivating it, should be allowed to participate in constructing financial infrastructure in different ways. As I have said, the infrastructure must take safety and stability into account.

Another concern is that payment systems and digital currency systems must consider their impact on the transmission of monetary policy, part of the broader concern of financial stability. That is, the transmission of monetary policy is an important channel for macroeconomic management in any economy. You can say that I have considered public spirit, security, stability, privacy and so on, but you have not considered the relationship with monetary policy and the degree of support for the transmission of monetary policy. This is also dangerous, because if monetary policy lacks a transmission mechanism, if the macroeconomy is out of balance, I think public infrastructure is not enough. Therefore, the building of financial infrastructure must be kept in mind when designing the system.

When I was in office, the PBOC and the payment industry jointly carried out the DC/EP research and development (R&D) scheme. The slash between DC and EP means both “and” and “or”, meaning digital currency (DC) and e-payments (EP) are compatible, not opposing each other. Suppliers should not label their own system as the only solution. Instead, multiple channels should be encouraged to research and compete in a manageable manner, in order to achieve a high-efficiency, low-cost and secure payment system.

One way to support currency stability and the transmission of monetary policy is to require any one unit of payment to be backed by 100% reserves. To a certain extent, we learned from the practice of Hong Kong. Before the handover of Hong Kong in 1997, there were two commercial banks issuing banknotes, HSBC and Standard Chartered Bank, and after 1997, Bank of China (HK) joined their ranks. The Hong Kong Monetary Authority, the de facto central bank, doesn’t issue notes directly. This is a little bit like PPP, where the central bank delegates note-issuing to commercial banks but still regulates by requiring them to submit $1 in return for a certificate of indebtedness per HK$7.8 issued. That is to say, unlike the “wildcat banks” of the 1920s and 1930s, issuing authorities are restrained.

The issues of so-called “anti-money laundering” and “counter-terrorism financing” must be also be considered according to global agreements and requirements. Obviously, money laundering and terrorism financing are threats to economic and social stability, and should be tackled with global coordination. The developers should bear public duties in mind, since the technology can be used both for good and for bad, just like how new biotechnology is designed for medical treatment and ecological conservation. But, if not monitored strictly, it risks being used as bioweaponry as well.

A bunch of DC/EP new technologies were first used on the “Dark Web” before being widely adopted by the public. Many media outlets, including Caixin, have covered the “darknet” where illicit transactions, tax evasion, money laundering, weapons trade, human trafficking and document forging are happening. On the darknet, transacting parties want to trade anonymously and without supervision. In view of this, the new systems must be very careful to comply with current requirements on anti-money laundering and counter-terrorism financing.

The last point is about cross-border payments, which has already been heavily influenced by the new e-payment and digital currency technologies. The technologies will surely influence cross-border payments as well, further facilitating it.

But cross-border payments are different from domestic payments in terms of what is needed. We mainly care about efficiency; the current system for cross-border payments is not efficient enough. At the same time, we are also aware that cross-border payment systems involve monetary policy sovereignty. A country’s macroeconomic policy mainly regulates the domestic economy, and cross-border payments could affect macroeconomic control and management.

Then there’s the question of financial stability. We know that emerging markets are suffering from instability, caused by devaluation and capital outflows. Argentina is the first case, starting last spring, and now it’s seen in Turkey, South Africa, India, Indonesia and Russia. Many countries have experienced it to different levels. So stability is also something that we should consider.

I mentioned that “stable coins” are pegged to sovereign currencies. Will they be pegged to something for cross-border payments as well? They will need something to be pegged to, something like the SDR, a mixed basket. Pegging to a single currency or gold is not enough. We should look into how to create a stable and accurate basket for payment systems.

Finally, we need a global coordinating mechanism for cross-border payments. As of now, there is no “global central bank.” Banks, especially some national central banks, have emphasized that their role is to serve the national economy and they have no responsibility to consider the spillover of its policies to other countries. However, especially since the financial crisis, we all know that monetary policy has spillover effects - fiscal policy as well, that is why we need global coordination.

Thus, if cross-border payments, electronic payments and digital currencies can take a big step forward in efficiency, there is still much research needed, including whether it will involve a global authority or coordination among global central banks to support this.

Besides coordination, we also have to prevent certain practices. Some nations quite like to impose financial sanctions that impact monetary policies and in turn affect e-payment and digital currencies. This is why we have much to consider before being able to set up cross-border payment systems.

Regulation is essential for financial technologies, because in the process of development there may be distortions to the market and disruption to society or too much short-term speculation. These phenomena have all happened, so they need supervision. Meanwhile, I want to stress that regulation must also evolve, as the development of technology might require much more from regulators. They must depend on technology much more than before, so more R&D is needed to support regulation.

Regulators must not be too oppressive and suppress the appearance of new technology. But they cannot become too loose and only take action when something bad has already happened, causing damage to society. We have had a taste of this tough balancing act already, as it is indeed hard to find the optimal equilibrium.

For regulators, they must renew their knowledge structure and their talents. Another thing regulators must consider is how many regulatory designs they want to have: Should there be a competition mechanism for the different regulatory designs? Or should we stick with one design once we are sure it is best?

But even if you have the best design now, you won’t be able to know how long it will stay the best. For any financial technological service, you have to consider how to set up an experimental environment for it, let it grow in it, while maintaining control over it.

Translated by Zhang Qizhi (

You've accessed an article available only to subscribers
Share this article
Open WeChat and scan the QR code