Bejoy Das Gupta, Chief Economist, eCurrency
The recently-concluded IMF/World Bank Spring Meetings in Washington DC brought together finance ministers, central bank governors and other senior officials from more than 180 countries to take stock of the global economy and discuss key policy and structural issues. This year, the IMF’s New Economy Forum at the Spring Meetings explored two key areas in financial innovation and fintech - Central Bank Digital Currencies (CBDCs) and the Economics of Data. The IMF could have chosen many aspects of fintech, but decided to focus on these two subjects, indicating their central importance to the authorities, multilateral institutions and the private sector.
CBDC Panel with Timothy Lane, Cecilia Skingsley, and moderator, Tommaso Mancini Griffoli
There were two sessions focusing on CBDCs at the New Economy Forum. The first was on “CBDC-Should Central Banks Issue Digital Currencies,” with the panelists being Timothy Lane (Deputy Governor, Bank of Canada) and Cecilia Skingsley (Deputy Governor, Riksbank), moderated by Tommaso Mancini Griffoli of the IMF. These central banks are at the forefront of the thinking and discussions on CBDCs. The panel was of the view that CBDCs were a very important issue, closely following climate change, given the rapid changes in technology and the financial system, with good possibility of being introduced within five years. While bank notes, which have been around for 150 years or so, are being displaced by commercial bank-provided digital currency, the central question was “how to provide a service, which was a trusted, safe, and efficient means of payment, which works for all in society.” While the precise form could evolve and change, CBDC would be a central bank liability transferable by digital means. Ideally, it should be cash-like by being non-interest bearing. While negative interest rates on
CBDCs were a theoretical possibility, taxing the public would limit its use while bearing interest could make them more attractive relative to bank deposits. Moreover, there should be privacy in transactions, while guarding against being used for illicit activities by limiting anonymity.
The panel was strongly supportive of the need to issue CBDCs as a broad public policy goal. The decline in cash in circulation to only 1% of GDP in Sweden as against 10% in the EU and 20% in Japan today was seen as a sign of things to come because of the convenience of digital payments and money in digital form. It was emphasized that while people could choose between cash and money in the bank today, it was viewed as essentially the same thing because the latter was protected by trust, regulation and deposit insurance. CBDC would allow this choice or “strategic complementarity” to be preserved in the future. Otherwise, there could be a vicious cycle, with households using less and less cash, retailers not accepting cash for payments, and banks not providing cash facilities.
Hence, the panel felt that there was a need for a CBDC, which people could hold and use as it replicated the positive features of cash in a digital form, while maintaining the functions of cash. Otherwise, the payments system would increasingly be dominated by a few large potentially foreign private providers of money, raising concerns of consumer protection, efficiency and stability. Regulation and supervision of a private network was not seen to be a substitute, because of the payments system being a natural monopoly as a result of economies of scale, network externalities and centralized settlement. Those wanting to continue to use cash would also be excluded. Moreover, there was the fundamental question of how well such a private system would work during times of crisis? As such, central banks did not want to take the chance. It was pointed out that people lost trust in the financial system in the 2008/09 crisis, and central banks did not want trust in the monetary system to be lost as well. Hence, the need for continued access to central bank money with monopoly of issuance and the choice with bank money to be preserved in the future. The challenge was to design a CBDC, which people would want to hold and use.
The use of CBDCs for helping facilitate cross-border transactions, which were slow, inefficient and expensive, was discussed. The Bank of Canada, Bank of England and Monetary Authority of Singapore are collaboratively studying cross-border CBDCs for wholesale payments and settlement, but these could also be extended to retail payments. The panel was of the view that these could be useful, but technology, policy, and regulatory differences would have to be resolved. Overall, it was a matter of mutual trust with regard to use of high-powered money, regulatory coordination and making payments systems interoperable. On the question of non-nationals being allowed to hold CBDCs, it was more of a political issue, although in principle they could. Meanwhile, technology was improving and moving towards instant settlement with countries such as Sweden considering the use of the ECB system. Another issue was countries with poor institutions, volatile exchange rates, high inflation and starting to dollarize moving to use of CBDCs issued by more stable countries such as e-krona or digital version of Canadian dollar. As such, any successful digital currency could potentially become global. However, there could be positive spillovers with neighboring countries also shifting to issuing CBDCs. Question for central banks is “whether we are providing a product as good as we possibly can...stable store of value and effective medium of exchange.” If not, “other version of money could come in,” such as dollarization or cryptocurrencies. In sum, competition is potentially good as it brings pressure on weak countries to improve and make their currencies more stable and trustworthy.
With respect to the risks in addition to the technical hurdles preventing central banks from immediately introducing CBDCs, the panel pointed to the need to consider consequences for the financial system. An appropriate design would guard against disruption of credit intermediation along with bank runs. While convertibility between central bank money and bank deposits underpinned public confidence and trust in the banking system, if the public rushed all at once to convert from bank deposits to central bank money, they could precipitate bank runs in unstable times. Accordingly, Timothy Lane pointed out that CBDCs should not be so unattractive that the public would not want to use them, but they should also not be so attractive that the public would not want to hold bank deposits. Such was the nature of the two-sided risk. He remarked that there had not been any systemic bank runs in Canada in more than a century, and individual bank runs have been movement of wholesale deposits from one bank to another rather than retail deposits from out of the banking system to cash. Cecilia Skingsley noted that we were already living in a world where we can have cyber-runs, because of the ability to move money from one bank to another or to money market funds using one’s iPhone, and the e-krona will not fundamentally worsen the situation. Moreover, banks provide differentiated services, including mortgages, other loans, savings products and so on.
Where would the new equilibrium end up depended in part on how banks would respond to CBDCs. Higher interest rates on deposit rates or squeezed margins? Or, raise lending rates to preserve margins, which could negatively impact credit extension. Timothy Lane noted that this was a consideration but not a primary worry in Canada. The Canadian banking system was already facing a fast-changing and competitive landscape, and has adapted to the changes in technology over time. Cecilia Skingsley was also in agreement that CBDCs should not make a difference, noting the challenges already emanating from fintech and other payment providers. The question was also raised whether the development of faster, cheaper, and more efficient and reliable RTGS payments systems negated the need for CBDCs. Sweden for example was thinking of shifting to the EBC’s new TIPS system, offering immediate settlement at low cost by cutting out the middlemen. Improved RTGS systems could also help in countries were cash usage was high. However, both panelists were of the view that only some of the need for CBDCs would be met, and there would still be a need for instant settlement using central bank money. With regard to the privacy of data under CBDCs to ensure that it is being used wisely, the matter was not really a question for central banks, but a broader political and legislative issue. Cecilia Skingsley liked the European approach, where individuals can give the rights to use data but could also revoke them.
The panel made the following observations during the Q&A. First, jeopardizing the effectiveness of monetary policy by displacing the role of banks in financial intermediation in a CBDC world was not a desirable outcome, and if it happened, central banks would have to recycle funds back into the banking system so that the footprint was unchanged relative to a pre-CBDC environment. Second, it did not make sense for a region to have a common currency for exports separate from individual currencies, and it was better to make them more trustworthy and functional. Third, tokenized CBDCs could make cross-border payments more efficient by being a catalyst for harmonization of regulations, but a consensus was needed. Fourth, with regard to the multiplicity of private bank-issued tokens, there would be a public sector role for establishing common standards, either through CBDCs or regulations. Fifth, reserve-backed e-money operations were not CBDCs because the governance was different, with backers of the e-money creating the backing and the trust rather than the central bank. Sixth, with regard to whether the economics or the technology came first when designing CBDCs, the technology was seen to be the tool, but could influence selection of economic attributes. Moreover,along with the economics and technology, a legal basis for supporting new forms of money was needed.
CBDC Panel with moderator, Sonja Davidovic, Marla Dukaran, Bejoy Das Gupta, and Adolfo Sarmiento
The second panel on “CBDC for Financial Inclusion – Design Considerations” included Bejoy Das Gupta (Chief Economist, eCurrency), Marla Dukharan (Chief Economist, Bitt), and Adolfo Sarmiento (Head of Economic Policy and Markets, Central Bank of Uruguay), with Sonja Davidovic of the IMF as the moderator, adopting an innovative case study format. The focus was on CBDC design consideration for a small economy with pegged exchange rates, a shallow and small financial market, and where cash is king with most transactions conducted in cash. Half the population is excluded from the formal financial sector. Given this backdrop, the question was how to design a CBDC to help expand access to those segments of the population excluded from the formal financial sector.
Marla Dukharan set the stage by pointing out how the context for CBDCs in emerging markets was very differentfrom the first panel, which focused primarily on the mature economies of Canada and Sweden. In the case of the Caribbean, geography played an important role, with the nations ranging from the 700 islands of the Bahamas to the sparsely populated Guyana with a large hinterland and the eight-nation Eastern Caribbean Currency Union. Distribution of cash was expensive, could take days, adding to the cost of living and doing business. As such, there was a powerful user case for CBDCs. With low financial inclusion and high mobile penetration, the solution was access to financial services through a mobile phone. People could have access to financial services through mobile means, which were secure, effective, and more cost-efficient than banks. People could use them for peer-to-peer transactions and exchange of money. These considerations made for a compelling user case for CBDCs.
With respect to designing CBDCs in a developing economy context, Bejoy Das Gupta emphasized that it should be a digital token, created and issued by the central bank, distributed through commercial bank and non-bank payment services providers, for use by the individual consumers. The key attributes of the tokenized CBDC would be interoperability or the ability to transact between different payment systems, with 24/7 access, and not being tied down to any denomination structure by making the digital tokens easily divisible. Moreover, it had to be fully secure, to preserve trust and financial stability. The token would also be for retail use and not for large wholesale transactions, the needs for which were met by the RTGS system. The tokenized CBDC would be a classic public private partnership, between the central bank as the issuer and private payment service providers as distributors. The central bank would in effect piggy back off the innovations of the private sector, resulting in cost-savings and efficiency gains.
Adolfo Sarmiento explained the thinking behind the design of the e-Peso pilot project, launched in November 2017. It was driven by the financial inclusion strategy and desire to help financial innovation, bring down the use of cash and offer a mobile means of payment. After first determining whether the central bank had the legal power to issue CBDCs, the idea was to provide digital money through the cell phone for youngsters, while preserving the use of physical cash by the older generation. In essence, fulfilling the payments system mandate of the central bank, while learning in a controlled environment how the technology worked, how the population made digital payments using the e-Peso, and assessing cyber risk and IT problems. There was a controlled amount of money and users, cash-in, e-bills, managing e-bill payments through wallets, using cell phones to move e-Pesos, and gathering micro data on how money was used for small payments. The pilot was successfully operated for six months, before cash-out and closure. Bejoy Das Gupta emphasized that it was very important when designing that cash and digital tokens operated side by side, and that it was a gradual process of rolling out CBDCs, starting with a pilot with several payment providers, before full economy-wide implementation.
The question when should central bank governors starting thinking about technology issues was then raised, especially when the IT staff in a typical central bank appeared to be enamored with blockchains and was pushing them hard. Marla Dukharan was of the view that for universal financial access, the need to move around digital cash seamlessly and easily, the banking system could not be used. Blockchains allowed for transaction and settlement at one shot. Hence, she felt that technology needed to be considered first in the design process. On technology, Bejoy Das Gupta had a different view, which is that the policy and design considerations should come first. It was very important to take into consideration monetary policy implications as well as no disintermediation of the banking system and maintaining financial stability when designing CBDCs. By enhancing financial inclusion, CBDCs could even help improve the transmission of monetary policy. Moreover, financial integrity was important, and in this regard the CBDC design could utilize the payment service providers to ensure that anti-money laundering (AML), know-your-customer (KYC) and combating the financing of terrorism (CFT) requirements were met by users.
In Das Gupta’s opinion, technology considerations should come at the end and the decision on which technology to use should be taken after arriving at the desirable attributes of the CBDC system. Moreover, blockchain was just one technology and there could be others, which could better meet the design attributes of CBDCs. In particular, whichever technology was chosen, it had to be able to meet the scale needs of CBDCs, be able to deliver finality of settlement, and be absolutely secure. Overall, the monetary policy, financial stability and IT groups in central banks should all be involved in the CBDC design, technology choice and implementation process. Adolfo Sarmiento was in agreement, indicating that many different groups needed to get on board and that design would be helped by an interactive process
Another question raised was that, given people frequently lacked proper identification, how to ensure that payments systems met identity requirements as part of the AML/CFT framework. Adolfo Sarmiento emphasized that technology would evolve and that the design needed to take this into account. Marla Dukharan pointed to the Singapore case, where KYC was kept in a centralized system, and which could be used by financial services providers, and emphasized that facial recognition could be utilized in e-wallets. Bejoy Das Gupta emphasized that biometric technology for proper identification was already available today, as evident for the successful Aadhaar program in India for a billion people. However, the main point was that the CBDC design should be such that the central bank, as the issuer, was sitting on the top of the pyramid, benefitting from the innovations of the payments service providers, through the public private partnership. In essence, technology was constantly changing, and meeting the AML/CFT/KYC requirements should be taken care off by the payments service providers and that the central bank need not worry about it.
Bejoy Das Gupta also brought up the anonymity versus privacy design issue. In this regard, the token CBDC should be designed in such a way that the anonymity features of cash were preserved so that users would want to use the instrument, while ensuring that they were not used for illicit purposes. The solution was that while the central bank had a dashboard which could see all the transactions, they were anonymous up to a certain threshold set by the political process, and high-value transactions beyond the threshold would not be anonymous. The central bank could ask for the data, if needed.
In the Q&A, Marla Dukharan mentioned that the publicly known CBDC issuance plans in the Caribbean included the Eastern Caribbean Currency Union and the Bahamas, and that the design called for running in parallel with notes and coins while seeking to reduce their use. With regard to risks, Adolfo Sarmiento spoke about cyber and AML risks, need to think about the velocity of money, security and anonymity design aspects of CBDCs, and the need to ensure the involvement of banks. Bejoy Das Gupta mentioned that the CBDC design needed to maximize benefits by bringing the non-banked into the formal financial system, which was as much as half the population in a country like Indonesia, while mitigating risks including not
disintermediating the banking system and preserving financial stability. On the pollution aspects of cellphones, which was needed for digital cash, they were already ubiquitous and proper disposal was broader than just a central bank issue.
In sum, the key takeaways are that momentum for retail CBDCs is building among central banks, pilots are on their way and that proper design is important to maximize benefits and guard against risks. Overall, thanks to the IMF for taking the lead for providing a forum for discussing the important topic of “why CBDCs and how to design them” during the Spring Meetings.
Herewith link to IMF’s New Economy Forum videos