David Andolfatto
The Bank of Canada (BoC) says it has been researching the idea of a central bank digital currency (CBDC) for several years to prepare for the future of money and interbank payments. A CBDC would allow individual Canadians, not just chartered banks, to open accounts with the BoC. I remain relatively agnostic on the proposal. It’s not essential at a retail level but I see merit in it at the wholesale level.
Most Canadian households have accounts at private sector banks. No Canadian household, however, has an account with the Bank of Canada, the nation’s central bank. Nor, for that matter, do most Canadian businesses. The only private sector agencies permitted to have accounts at the Bank of Canada are chartered banks. These accounts, however, are not used to facilitate transactions. Credit balances in these accounts (which may be negative) are called settlement balances. They are used primarily to settle the interbank obligations that arise at the end of each day as a byproduct of clearing household, business and government payment requests. Net surplus/deficit positions across banks are settled in a process that involves banks either accessing the market for settlement balances (an interbank market) or by depositing/borrowing settlement balances to/from the Bank of Canada. The deposit and lending rates at the Bank of Canada’s deposit and lending facilities are used to implement interest rate policy.
A central bank digital currency (CBDC) is a proposal to grant households and businesses the right to open transaction accounts with the Bank of Canada. Since every household and business already has the right to open transaction accounts with Canadian chartered banks, what is the case for adding a retail-payment division to a Crown corporation specialized in clearing and settling payments between banks?
Existing issues
Clearing and settling payments at its core boils down to an exercise in messaging and bookkeeping. A payment is initiated when a message is sent with an instruction to debit one account and credit another. Imagine having to design a payment system from scratch. What could be simpler than having a centralized ledger operated by a regulated entity charged with the responsibility for validating and processing messages to debit and credit money accounts? This sounds like basic public infrastructure and it could be financed as such. As CBDC balances are direct liabilities of the central bank, deposit insurance would be unnecessary. To address concerns over fairness, minimum balance requirements and user fees applied to small-value accounts could be abolished. Protocols could be put in place to address concerns over data privacy and ownership. As well, a CBDC employing the most up-to-date messaging systems with real-time payments functionality would be expected. As an open-access facility, a CBDC would permit fintech firms to bypass conventional banks and their monopoly access to settlement balances, thereby cutting out an unnecessary layer of intermediation. In short, a CBDC would be both efficient and fair.
Alas, payment systems are not developed from scratch. In Canada, they’ve evolved over time to form an increasingly integrated patchwork of networks — the design of which has been determined by a host of factors, including available technology, competition and political pressure. Dingle (2002), for example, describes in some detail the economic, political, and technological factors leading to the establishment and evolution of the Canadian Payments Association (also known as Payments Canada), a private-public sector organization responsible for overseeing the Canadian payments system. The dominant payment rails in Canada today are provided by a coalition of banks (via Interac) and the two major credit card companies (each of which provide debit and credit cards). In terms of their debit card business, credit card companies operate more like messaging systems conducting debit/credit operations that occur on customer bank accounts rather than on their own accounts. An emerging industry of digital wallet providers are competing to offer similar services. Other non-bank payment service providers (PSPs) — for example PayPal — offer clients transaction accounts with uninsured money balances pegged to the Canadian dollar. A more recent incarnation of this model is emerging in the form of so-called stablecoins. These latter entities look very much like old-style money market funds with chequing privileges. Unlike bank accounts, which are presently insured up to $100,000 by the Canadian Deposit Insurance Corporation (CDIC), non-bank accounts are backed entirely by the securities and other assets they own. Not surprisingly, the emergence of uninsured non-banks operating in the payment system has raised concerns related to consumer protection, data privacy and ownership, and financial stability.
Given what we have in Canada today and on the near horizon, how far away are we relative to the hypothetical CBDC ideal sketched above? The answer to this question obviously depends on the metric employed. My own view is that we’re not that far off and that the deficiencies people commonly cite are better addressed directly through legislation, rather than indirectly through the provision of a retail-level CBDC. Having said this, I remain largely agnostic on the proposal, although I do see merit in the idea of a wholesale CBDC.
Let me explain how I came to these conclusions.
To begin, a vast majority of Canadians are connected to the banking system. Transaction accounts living in different banks are connected through Interac, a coalition of private banks that operate an interbank payment system. Thus, to a first approximation, Canadians are already connected to each other on a consolidated bank ledger. A centralized ledger provided by a CBDC for retail use would therefore be largely redundant (though, some redundancy made be desired). There is the question of why fintech firms must be forced to operate through incumbent banks. One option would be to grant such firms the ability to open accounts directly with the Bank of Canada and to participate in the large value transfer system (LVTS) through a “narrow bank” charter. Doing so would be tantamount to offering a wholesale version of a CBDC.
I mentioned that most Canadians have bank accounts. A small percentage of households, however, do not. It seems unlikely, however, that offering yet another online payment service — even a free one — would induce these individuals to suddenly access the digital payment system. No-frill, no-fee payment services already exist in Canada, for example, via Tangerine. Any effort to encourage or otherwise help those without bank accounts to access the banking system could be accomplished using the products presently available.
Potential roadblocks
As mentioned above, money balances held in CBDC accounts would be fully insured. Because any CBDC constitutes fiat currency, it represents a claim against itself. This is in contrast to the digital currency produced by banks, which represent legal claims against fiat currency, as well as other bank assets in the event of bankruptcy. Whether this legal distinction matters in today’s world, however, is debatable. The Canadian banking system is a private-public enterprise. The government interacts with private banks through the Bank of Canada, the CDIC and several other regulatory agencies. Because bank accounts are federally insured up to $100,000 per account, it seems unlikely that households and small businesses would find a CBDC attractive in terms of relative safety. The same consideration is not true of money held in non-bank accounts, of course. But people using uninsured accounts already have the option of holding transaction balances in CDIC-insured accounts. Why would the added option of a CBDC have any material impact in this regard?
Even if people are not concerned with the safety of their deposits, they may be concerned about the security of the personal information associated with their accounts. In addition, while account managers are obligated to respect privacy laws, they are inclined to view some data – such as purchase and credit histories — as proprietary information to use, and possible sell, to interested third parties. While a CBDC could be designed in a way that transfers data rights back to users, doing so would not necessarily compel banks and PSPs to follow suit. If such an outcome is deemed to be in the social interest, then legislation — say, similar to the PSD2 regulation in Europe — seems like a more direct way to achieve it.
Let me next ask how Canada has kept up with best practices in the payments system. The agency responsible for operating and developing the Canadian payment system is Payments Canada, a private-public consortium overseen by the Ministry of Finance and established by the Canadian Payments Act in 1980. Dingle (2003) describes how Payments Canada managed a successful transition from paper-based payments to electronic payments in the 1980s and 1990s. The two major payment systems owned and operated by Payments Canada today are its large value transfer system (LVTS) and its automated clearing and settlement system (ACSS). While these technologies appear to be lagging those available in some other jurisdictions, a new high-value payments system called Lynx is soon scheduled to replace the LVTS and a new real-time payments system called the Real-Time Rail (RTR) — a 24/7/365 real-time payment system — is scheduled to augment the ACSS (a retail batch payments system, which itself will be upgraded). Associated with these changes will be the incorporation of modern rules and standards, including the ISO 20022 global payment messaging standard.
How rapidly might a Crown corporation have kept up with rapidly evolving technologies and protocols is an interesting question with no obvious answer. It seems highly likely that any such organization would have in any case consulted and co-ordinated with members of Payments Canada, of which the Bank of Canada is presently a member. It is perhaps charitable, though not too charitable, to suggest that Payments Canada has demonstrated a capacity for remaining at or near the technological frontier in this regard. If adoption appears slow because of regulatory hurdles, then work to remove the regulatory hurdles. If adoption appears slow because market power impedes competition, then work to remove barriers to entry and promote competition. While a CBDC could be used a way to promote competition, it is not essential to achieve the desired outcome.
If factors such as access, safety and speed are not considered major defects of the current system, then what is? A perennial issue concerns the (allegedly excessive) fees that card companies and their acquirers charge merchants, especially smaller businesses, which do not have the bargaining power to negotiate more favourable terms. It has been suggested that a CBDC would lower these fees by providing a cheaper alternative for processing payments. This conclusion, however, may not be warranted. Let me explain.
The business model employed by card companies is to reward cardholders for spending their money and recoup these and other expenses through merchant fees. Consumers appear to value this arrangement, but merchants do not. Because consumers love reward and cash-back programs, they demand the opportunity to pay with the cards offering the best rewards. Merchants have little choice but to accept these forms of payment or risk losing market share. Moreover, the terms that card companies impose on merchants often restrict them from applying payment-contingent pricing policies, an example of which is the so-called “honour all cards” rule. The effect these restrictions have is to encourage consumers to select the payment option with the highest private reward — not necessarily the one with the lowest overall social cost. Higher costs must, of course, be absorbed along other dimensions, including higher product prices.
Given the business model described above, how might an even zero user-cost CBDC attract consumers? Cardholders already have access to low-cost payment methods. A CBDC may potentially offer consumers a relatively high interest rate on their deposits, but this is a reward for saving. Card companies reward consumers for spending. While merchants would no doubt be happy if consumers were to switch en masse to CBDC cards, there’d be no point in introducing the product — at least for this purpose — if consumers are not willing to part with their beloved reward programs. So, if the reward programs are the problem (I am not passing judgment here), then abolish the practice as a condition for obtaining a business licence in this sector.
Four other issues
A CBDC is also promoted on non-payment-related grounds. I list the four most-commonly made arguments, each followed by a short rebuttal.
First, should physical cash disappear, a CBDC could provide its digital equivalent. If physical cash disappears, however, it will be because nobody wants it. It makes no sense to provide a digital equivalent of something nobody wants.
Second, a CBDC would disintermediate banks (this is considered a feature, not a bug by some proponents). I very much doubt this would happen, especially if the CBDC policy interest rate was set below the interest banks earn on settlement balances. To begin, banks would surely compete to retain their deposit funding, squeezing profit margins, but not necessarily diminishing lending activity (Andolfatto, 2021). History also shows how resilient Canadian banks are in this regard. For example, banks lost a cheap source of funding in 1935 when they were prohibited from issuing banknotes, yet they managed to prosper in spite of the restriction.
Third, a CBDC is touted by some as a way to overcome the zero-lower-bound constraint on central bank interest rate policy. The desirability of such a policy is debatable But even if it is desirable, there is nothing in principle to prevent negative interest rates from being offered at the Bank of Canada’s deposit and lending facilities. I do not think small denomination cash prevents a negative interest rate policy.
Fourth, some have expressed fears that global competition — from other central banks offering CBDCs or from large online social networks — may hamper Canada’s monetary sovereignty. This fear seems somewhat exaggerated. While there may be some currency substitution, as long as the Canadian government requires taxes to be paid in Canadian dollars, the demand for the product is not likely to disappear. In any case, it’s not clear how a Canadian CBDC is supposed to discourage Canadians from holding transaction balances in foreign currencies, or encourage foreigners to hold transaction balances in Canadian dollars. This is, of course, assuming that CBDCs will be no less-constrained than private banks and PSPs presently are in terms of restricting the frequency and value of money transfers as a part of fraud prevention.
Conclusion
A retail CBDC is, in principle, an attractive proposition. But given the system currently in place and the prospects for its near-term evolution, a retail CBDC on its own is, in my opinion, not an essential initiative at this point in time. For consumers, a retail CBDC would mostly replicate what they already have available. As such, the initiative is not likely to attract business away from private-sector PSPs or serve to discipline private-sector pricing protocols and rewards programs. For a CBDC to be successful in this regard, legislation (or moral suasion) designed to alter private-sector marketing behaviour is likely required. But if such legislation were forthcoming, the rationale for a retail CBDC is even further diminished. On the other hand, a wholesale CBDC (together with legislation governing pricing protocols) seems like the most straightforward way to promote competition and fairness in the Canadian payments system. While I see no reason why a CBDC could not work in principle, I also do not see why it is essential in practice. It probably makes more sense to let the Bank of Canada focus on its core competencies — monetary policy, regulation and wholesale payments — and let a regulated private sector manage retail payments.
Acknowledgements: I would like to thank John Murray, J.P. Koning, Chris Waller, and Werner Anteweiler for several helpful conversations that have contributed to the views I express above. This does not of course mean they should be held responsible for my conclusions. The views and opinions expressed above are my own and should not be attributed to the Federal Reserve Bank of St. Louis or the Federal Reserve System.
