Interview with Financial Times
Interview with Fabio Panetta, Member of the Executive Board of the ECB, conducted by Martin Arnold on 14 June 2021
20 June 2021
You are getting close now to a decision on whether to pursue further work on the digital euro. You recently completed your public consultation on this, in which the biggest concerns were about privacy. So how do you address people’s concerns about privacy and still do all the necessary checks to make sure that it’s not abused for money laundering, tax evasion, and all of those things?
If the central bank gets involved in digital payments, privacy is going to be better protected. Why? Because we’re not like private companies. We have no commercial interest in storing, managing or monetising the data of users of a digital means of payment. We’re not a profit-maximising institution, we work in the interest of citizens. So we’re a different animal than private service providers. This also emerged from the public consultation. People feel safer if their information is in the hands of the central bank – a public institution – than if it is in the hands of private companies. Second, there are many ways in which we can protect confidential data while allowing the checks foreseen by law to avoid illicit transactions, such as those linked to money laundering, the financing of terrorism or tax evasion.
How would that work?
First of all, we could segregate the data. Suppose that I have to give you money. I’m the payer and you’re the payee. I would go to my bank, which would know that I’m making a payment and would transmit a code to the payment operator. The payment operator would transfer the payment between one code and another code. It would not know the identity of the payer behind the one code and the payee behind the other. So the payment will go through, but nobody in the payment chain would have access to all the information. But this is only one example. We could use cryptographic codes. It could be possible, for example, to make offline payments for small amounts, in which no data are recorded outside the wallets of the payer and payee. One could imagine that for payments of small amounts – €70 or €100 – you could have offline payments without a connection.
But don’t you need to link back to the identity if you’re going to restrict people to a certain amount?
For these amounts, one can also imagine that you don’t do that. Why do we want to have all payments traceable? Because there are issues in terms of money laundering, the financing of terrorism and tax evasion. This risk is much lower for small transactions, as long as they are not used to split a larger payment into many smaller ones. We addressed this in past experimentation by introducing “anonymity vouchers”, making it possible to anonymously transfer a limited amount of digital euro over a defined period of time. So the concept is that controls should be cost-effective. For very small amounts, we could permit truly anonymous payments, but in general, confidentiality and privacy are different from anonymity. Full anonymity must be considered very carefully, because there is a trade-off between guaranteeing full anonymity and guaranteeing compliance with fundamental regulations in areas such as anti-money laundering, combating the financing of terrorism and tax evasion. And let’s not forget that citizens will still be able to use cash, which guarantees anonymity. There are many things we can do, with the help of technology, so people feel safe about how their data are used, and at the same time, so a payment can be reconstructed ex post if the police want to assess whether there’s been any illicit activity, any crime going on. But that would not be in line with the incentives of private intermediaries, which are interested in the commercial value of transaction data. We’ve done some preliminary experimentation on how to safeguard confidentiality with national central banks, which will be published. So we’re discussing possible ways to guarantee privacy. Why are we doing this? For obvious reasons and because the message that emerged from the consultation, as you said before, is very clear. Privacy is a top priority for users.
What about people who say that, while the ECB may not have commercial interests, you are a public sector body? So the idea of the government spying on people by looking at what they’re doing with their money, isn’t that also a concern for people?
I start from the assumption that people should be able to rely on a public institution, and we’ll make sure to set up governance structures to avoid any possible abuse of data. We’ll act within the scope of European legislation, which is the most advanced worldwide in protecting data, with an independent data protection supervisor that we will be interacting with.
Can you tell me a bit more about the experimentation? What does that entail?
The experimentation we’ve done so far is to get a preliminary sense of the pros and cons of different technologies and the limits of, for example, handling payments while safeguarding confidentiality. In this preliminary phase, we’ve organised four work streams in which we’ve tested the possibility of running a digital euro with a centralised system, a decentralised one, a mixture of the two and with offline payments. If the Governing Council gives us the green light in July, we’ll start a formal investigation phase focusing on the design of a digital euro. After two years, we’ll get back to the Governing Council, and in the meantime we’ll interact with other European authorities and institutions – the Parliament, the Commission, the Council, the Eurogroup – all those who are involved, because the digital euro will require legislative changes. So at the end of these two years, ideally, we would have more clarity on the steps which would be necessary to issue a digital euro, if the decision were taken to launch it. Then we expect to have, by and large, three years to be able to implement what we have decided on, by working together with the technology providers and banks on the end user features of a digital euro so it could be integrated into the services that they’re already offering to their customers.
Why do you need legislation?
Because, for now, a digital euro is not explicitly foreseen in the European legislation. There are different ways in which you could define the legal basis for a digital euro, depending on its characteristics. Existing legislation may also need to be adjusted to cater for a digital euro. For example, to allow the anti-money laundering authorities to have powers to verify, ex post, not only transactions through bank accounts, but also through the digital euro.
Let’s take a slight step back. There are still some people who are scratching their heads and saying, what is the problem that you’re trying to solve with this project? What is the point of the digital euro?
First of all, for many centuries, from ancient Greece to the Roman Empire and Charlemagne, the sovereign has always offered money to citizens – sovereign money, which is the ultimate reserve of value for citizens. We’re doing the same. We intend to continue to do so. So why do we need a digital euro for this? There are two reasons. First, people are paying more and more digitally, and less and less with our current means of payment – cash. And second, people are buying more and more online, and with e-commerce it is relatively difficult to pay with cash. So people are using the means of payment backed by the central bank less and less. We’re moving into a digital era, so by introducing a digital euro we would be changing how people can access our balance sheet and use our means of payment.
So it is the decline of cash that you are responding to? But you can pay digitally with electronic money, you don’t need a digital euro to do so. I can already buy pretty much anything I like sitting here with my mobile phone.
First of all, there is no digital means of payment that you can use everywhere in the euro area, from Finland to Cyprus.
What about a Visa card?
You can’t use it everywhere. As you know, even here in Frankfurt you can’t use Visa or American Express in every shop. And it can be expensive to use them. Second, I think that this is not just about preserving the role of the instrument of the central bank. There are many other reasons. First, we would offer a means of payment that is risk-free. I think that people do have the right to access the balance sheet of the central bank, which is the only risk-free institution, even less risky than the sovereign. One could imagine that, in most cases, liabilities of commercial banks are pretty similar to the liability of the central bank. But there are situations when differences do matter. You remember in the financial crisis when banks stopped trusting each other. So in that situation, the liability of a commercial bank was not perceived by citizens or by banks themselves as riskless. So you must have a riskless means of payment, a riskless financial instrument in the economy. And this is the liability of the central bank.
Are you also worried about the threat from cryptocurrencies and other central bank currencies?
This is not why we started this analysis and will possibly start a project, but of course there is the potential threat that could come from others issuing a digital means of payment. This could be a so-called global stablecoin or another sovereign providing a digital means of payment. And if people do want to pay digitally and we don’t offer them a digital means of payment, somebody else would do that. So I can give you many, many reasons. And by the way, it’s not just about payments. This, I think, is a historical change that will not only change the way the financial system works. This will change the attitude of citizens towards digital instruments. I think this is a fundamental change to how payments will function in the future, both for the financial system and for society at large. And it is generating huge interest.
Do you think the end consumer is going to notice a big difference? Because legally they will have a claim directly on the central bank. But for all intents and purposes, it’s just going to be numbers on a screen, the same way that their digital deposits are shown on a banking app.
Most of the time I would agree with you, but not always. The great financial crisis was a very big wakeup call. It’s not true that people don’t differentiate. Most of the time, they don’t need to differentiate between different means of payment, among liabilities of different issuers of different means of payment. But there are situations in which people do understand very well. They optimise. They don’t care about whether they have bank deposits or cash, unless the bank deposit becomes risky. In that case, they seem to understand the difference very well and they seem to differentiate very carefully.
Some people imagine that the central bank is going to be operating these digital euro accounts directly. But is that really how it’s going to work? You’re more likely to outsource the management and the operation of this to the commercial banks and some fintechs, right?
It’s theoretically possible that the central bank would handle the accounts of individual citizens. But it’s extremely unlikely that this would be the choice we would make. Why? First of all, there are currently, I believe, around two million or more bank employees in Europe, dealing with 400 million customers. There’s no way that the limited number of employees at the central bank could do this. So even if we were crazy enough to embark on such a project, we could not do it. We’re very productive, but not that presumptuous. Second, we don’t want to use the digital euro to change the structure of the financial system or to destabilise the functioning of the financial sector. Banks already provide citizens with a large number of services, and in the future they would add access to the digital euro as one additional service to build a business model with the “digital euro inside”. Banks are also much better than central banks at onboarding, and at know-your-customer and anti-money laundering checks. We could not do it. So even if it is theoretically possible, it’s very unlikely. What we want is for the digital euro to be a sort of raw material that we would hand over to banks to provide the services they are providing now, plus access to the digital euro, in the same way they’re already providing access to cash.
Well it would be intermediaries that are involved in the provision of payment services. They would need to be regulated and supervised intermediaries. So not only banks.
Is there a danger of crowding out innovation by the private sector?
We’re a very peculiar supplier of services, because in general, private service providers want to take market share from competitors. We’re not that type of provider. What we want to do is negate the risk that the digital euro would crowd out banks and innovation in payments. So we’ll take care of that. I think the digital euro would be a source of innovation, not the opposite. And we’ll not crowd out banks, because this will be a public good provided by the central bank to foster innovation, progress and efficiency in the financial system.
That must be one of your primary concerns: to avoid destabilising the banking system in any way or facilitating runs on banks in crisis times?
Absolutely. We’re reflecting on how to avoid that. What we want to offer is a means of payment, not a form of investment. To avoid crowding out banks, we are discussing possible alternatives. One possibility is to put a cap on the amount of digital euro that individual users could hold. So you can hold a maximum of, say, €3,000, but not more. Any money in excess of that would have to be transferred into a bank account. Another possibility is that you can hold as much as you want, but beyond a certain threshold, you would face a financial disincentive. Up to €3,000 euro, amounts held in digital euro would never be treated less favourably than cash, they would never be remunerated at interest rates below zero. But if for some reason you wanted to hold more than that, then you would face a financial disincentive that would discourage you from doing so. Why would we do that? Because if people could hold an unlimited amount of digital euro, they could shift all their deposits from bank accounts into digital euro, especially in the run-up to a crisis, potentially precipitating a banking crisis. Which would destabilise the financial system. By the way, these proposals which originate from the ECB are now being discussed quite widely around the globe.
Are you leaning towards one or the other of those?
No, we will take that decision after consulting with stakeholders, citizens, merchants and banks, and there are implications and pros and cons in either method. For example, the cap may be more powerful. You cannot hold more than €3,000 full stop. Right. But then we would have to deal with a number of complications. Suppose that you have €2,800 in your digital euro account and I owe you €500. If I make a payment to you, what happens to the €300 that exceeds the threshold, the cap of €3,000? Either the payment cannot go through, which would be disastrous as it would create uncertainty about the completion of a payment, or a waterfall mechanism would have to be set up to transfer the rest to a bank account. But that would mean obliging you to have a bank account, which is at odds with the concept of financial inclusion, which is one of our objectives. Access would need to be provided to basic bank accounts that are free of charge. Tiered remuneration is of course more market friendly. But is it powerful enough? In a crisis, it could be very difficult to have a financial disincentive that is large enough to cause you not to seek security if you fear that your bank could go under and that you could be bailed in. So you would need to bring down the remuneration of digital euro in excess of a threshold when there are concerns about a financial crisis. But then that would be a signal that would need to be managed. So, you see that there are solutions, but each of them – cap or tiered remuneration – has advantages and disadvantages.
Is a digital euro going to use blockchain, or distributed ledger technology?
In our tests, we have experimented with two systems. One is a centralised system known as TIPS. We have used this to analyse the functioning of a theoretical digital euro. How many transactions can be done per second using that infrastructure? And how much time does it take to complete each transaction? We had excellent results with TIPS, suggesting we could handle hundreds of millions of transactions per day, and the completion time for payments was negligible. Then we have also looked at distributed ledger technology (DLT). While there are types of DLT which are used by commercial banks for financial transactions, there’s no experience of DLT that could serve the needs of 400 million customers. And we’ve also experimented with a combination of the two, because there’s no reason why one should exclude the other. We could have a centralised system with different nodes, and then these different nodes would interact with the centralised base enabling you to multiply the power of the system. But this is a question for the IT experts.
Regulation of cryptocurrencies and stablecoins, how do you see that? Do you think that stablecoins need to be more tightly regulated?
We still have unstable coins. To become stablecoins, a necessary although not sufficient condition is that they need to be regulated and supervised. And obviously, the coins have a number of risks for individual consumers and for the financial system at large. So what is the mechanism behind the so-called stable or unstable coins? The stablecoin issuer sells the coin to you, then uses the proceeds from the sale to invest in reserve assets. But those reserve assets do not have the certainty or stability of their value. They may be low-risk assets, but they are not risk-free assets. It is also uncertain whether the issuer holds an adequate amount of such assets to back the value of coins in circulation. Then there is the possibility of the value of the asset changing and the value of the coin changing. In periods of financial tension, there could be an expectation that the value of the coin would change significantly because of the value of the asset, and then there could be a risk of people going to the issuer and withdrawing their money. That is the mechanism that leads to a run. But unlike in the case of banks, there is currently no supervision. There is no deposit guarantee scheme and there is no emergency liquidity provision by the central bank, because the central bank cannot commit public money to save entities that are not subject to adequate prudential requirements, meaning that there is an inherent instability in the function of these coins – and for this reason they are still unstable coins. Once we have regulation and oversight, it is possible they might become more deserving of the name “stablecoins”. But only the money issued by the central bank is truly stable. Europe is at the forefront of regulation in this field. The European Commission’s legislative proposal for a regulation on Markets in Crypto-Assets (MiCA) sets out a number of mechanisms to reduce the risks to consumers and to financial stability. And the message sent is that until the regulation is in place, the presumption would be that no stablecoin should start operating in the euro area. At the same time, we are looking to adopt our oversight framework. We are currently discussing our new Eurosystem oversight framework for payment instruments – the PISA framework – and it will be adopted to include oversight of stablecoins. Once this is done, then we will discuss the possible introduction of such coins.
What about cryptocurrencies – bitcoin and the like? How are you managing them at the moment? Because the regulation framework isn’t in place yet.
There is no formal mechanism that we can use to prevent any investor in the euro area from buying crypto-assets. They are not currencies; they are not money. The risks they bear are very high. I think there should be careful regulation. MiCA introduces regulation for crypto-assets. Our PISA framework will also give us the possibility to conduct oversight of crypto-assets.
What does this mean, oversight of crypto-assets?
It’s very difficult to regulate them, to oversee them, because there is no responsible legal entity. They are decentralised. They could be in China. They could be in Switzerland or in South America. They could be anywhere. But to the extent that intermediaries are involved in the supply of those crypto-assets, then we would have regulation and oversight in place.
In my view, crypto-assets are very dangerous animals. They are not money, they are contracts that are perfectly fine for gamblers. We should try to protect consumers as much as possible. And we know from recent experience that those crypto-assets are largely used for criminal activities. We also have the issue of energy consumption. Bitcoin mining, for example, uses a huge amount of energy. This is not sustainable.
When will this supervision start? When is PISA coming into force?
By the end of this year, I hope. But we also need the MiCA regulation, because it is complementary to our PISA framework.
Let’s talk about monetary policy and the implications for that. You talked about imposing negative interest rates on digital euro to disincentivise people from holding too much of it. Could it be a way to make interest rates even more negative?
No, because first of all, we’re not planning to withdraw cash. The reason why there is an effective lower bound for monetary policy is that if you bring interest rates too low, people would turn to cash, as the cost of storing cash would become lower than the cost of holding a non-cash instrument.
We’re not introducing the digital euro because we want to change the way we implement monetary policy in any way. We’re not planning to withdraw cash. So we would never use the digital euro for monetary policy, and we would never use it to impose deeply negative interest rates by design, because we’ll continue to supply cash.
Have you got any idea of the cost of launching the digital euro project?
Currently we pay for the costs of producing banknotes and end users are not charged, but we are not losing money. Quite the contrary, because we earn seigniorage – which results from the difference between the face value of money and the cost of issuing it. Likewise, with the digital euro we won’t charge end users and we will incur some costs, but we won’t lose money. Seigniorage means we’ll have more liabilities and more assets, and we’ll make money out of that investment under normal market conditions. As part of the investigation phase, we’ll discuss the overall architecture and how the various stakeholders involved in the transactions charge each other, but citizens paying with digital euro won’t be charged. And I don’t think that the cost of issuing a digital euro will be large. We already have experience with wholesale and retail payments. Suppose that we use TIPS – the technology is already in place. And national central banks already use DLT. So I don’t expect the cost to be prohibitive. And the proceeds would be very large, because that’s a way of maintaining the size of the balance sheet, which means that you have more liabilities and more assets. This means that you’ll earn more seigniorage.
Will intermediaries make money on it?
The ECB as the issuer will not charge users. Intermediaries will offer services with digital euro inside to cover their costs. This is a complicated issue that will have to be addressed: the interaction between banks, intermediaries and customers. And competition has to be ensured to limit the fees and charges for customers.
How will this change the world of finance, the world of money for the consumer out there?
This would be a very powerful push for digitalisation. It would allow everybody to have access to a safe, risk-free, cost-free means of payment in the digital era. We’re currently still in a period of transition, but 20 years from now, everybody will be using digital instruments. The digital euro and cash could co-exist, but I’m sure that everybody would be using digital means of payment. You can build programmable products at lower cost. For instance, users could decide to allow automatic payments for routine transactions, such as paying bills, using a toll road, going to the cinema or parking.
So is this just for the eurozone or will it be available to people outside the eurozone? And if so, is there a danger you erode the monetary sovereignty of smaller countries?
There is indeed such a danger, and that is why we’re considering which potential users should have access to digital euro. Within the euro area, €3,000 would already be above the cash requirements of most people today. But in some countries, including some not far from us, with lower GDP per capita, €3,000 in digital euro form could be too large an amount, as it could trigger financial instability by depleting bank deposits. We’ll have to reflect very carefully on access and limits for foreign users. Of course, foreigners that come to the euro area as tourists should have access to digital euro. But here too, we would use limits.
Then there is another and more positive international dimension: the possibility of doing cross-border payments. We’re already cooperating at the international level. The advantages of introducing central bank digital currencies would be maximised if we made them interoperable, as they could make cross-border payments more efficient and much cheaper.