When £1 is not £1

  • A large number of central banks are currently researching and discussing the introduction of CBDCs.
  • The reasons for this and the form CBDCs will take, vary from country to country.
  • One risk that is less studied is what could cause the value of a CBDC to diverge from that of existing money.

At the moment, a large number of central banks, both in advanced and in emerging economies, are researching and discussing the introduction of so-called Central Bank Digital Currencies (CBDCs). [1]

The reasons for this vary. In some countries, e.g., Sweden, the central bank is concerned that the use of cash will diminish to a point where the general public no longer has access to central bank money. Should this occur, the Riksbank fears that it could have negative effects on the economy and might, in a crisis, cause problems in the payments system. In other countries, there is a concern that someone else (usually China) might be first with a CBDC and could therefore set the rules and control the infrastructure for this type of currency. A third reason is concern about cryptocurrencies and that these could come to play a major economic role, which would limit the capacity of central banks to conduct monetary policy. This worry was particularly noticeable when Meta (Facebook) was announcing its projected digital currency, the Libra, although this never took off.

The exact design of CBDCs is still not entirely clear and will almost certainly vary from country to country. Central banks continue to research and discuss different models, individually and collectively, not least through the Bank for International Settlements (www.bis.org).

There is, however, one aspect of CBDCs which seems to have been discussed less, although it has not been entirely ignored. That is a situation where a unit of the CBDC may not be worth the same as an equivalent unit of cash or commercial bank deposit. (For the sake of brevity and convenience, the latter will here be bundled under the term ‘legacy currency’.) In other words, if £1 (CBDC) ≠ £1 (legacy currency).

There are four aspects to this. The value divergence can be intentional or unintentional; and the CBDC can be worth more or worth less than its legacy currency equivalence.

The issue of an intentional divergence between a CBDC and the legacy currency has been covered in some academic literature.[2] This primarily covers the case where policy interest rates are at or near the effective lower bound (ELB) and the central bank is attempting to ease monetary policy further. One way to do so would be to impose sharply negative interest rates on the CBDC in order to discourage saving and encourage spending. However, since this could lead to a flight from the CBDC to legacy currency, the policy would also involve depreciating the legacy currency vis-à-vis the CBDC at a pace to make switching between the two irrelevant. 

This is not a policy without risk or complications. Market actors would have to be convinced that parity will eventually be restored. If not, it is possible to see cash becoming more attractive and commanding a premium. This would then necessitate even deeper negative interest rates on CBDCs in order to provide the required stimulus, potentially causing a vicious circle of value divergence to develop. Further, contracts have to be written so that debts cannot be repaid in depreciated legacy currency – which probably means making the CBDC the legal numeraire. There are also questions over whether retailers would pass on the currency divergence to clients, or assume the exchange rate risk themselves? Finally, there is also the political risk, where markets, having once experienced an intentional currency divergence, could fear that it might be repeated in the future. Nevertheless, it is possible to see such a policy attempted in a situation where more conventional monetary policy instruments seem powerless.

By contrast, it is difficult to see why a central bank would intentionally let its CBDC depreciate relative to legacy currency. It might possibly occur in order to discourage a flight from legacy currency in a bank run.

The situation where a legacy currency could unintentionally depreciate vis-à-vis the CBDC (i.e., an appreciating CBDC) is easy to deal with. The central bank can simply announce that it will at any time exchange legacy currency for CBDC at par.

However, what if the situation is that the value divergence is not only unintentional, but also involves the CBDC depreciating against legacy currency? 

Why and how could this happen? There are a few possibilities. One would in fact be in the situation described above, where monetary policy involves deeply negative interest rates on CBDCs, coupled with a ‘managed decline’ in the legacy currency/CBDC exchange rate. In particular, if retailers in this situation were to protect consumers from the impact of depreciating cash, that could counter the central bank’s monetary policy and make cash more attractive relative to a CBDC. In other words, Gresham’s Law would go into reverse and ‘good money’ would drive out ‘bad’. While that ought only to happen if the exchange rate between the two currencies is floating, history shows that fixed exchange rates have a habit of breaking down when they diverge too far from economic fundamentals. 

Another possible way in which a legacy currency, in this case, in particular cash, could develop a premium, would be if market actors had a reason to fear that their CBDC holdings could be reduced by the state for any reason – not just negative interest rates, but potentially as a tax or in expropriation.

A third possibility is where, for some reason, legacy currency, notably bank deposits, are perceived to be safer that ‘state money’. This is admittedly rather far-fetched, since bank deposits usually are denominated in the state currency. However, the Greek crises of the early 2010s could conceivably have seen such a situation develop, where market actors abandon both cash and CBDCs – although this would probably have to be accompanied by a shift from domestic-denominated bank accounts into foreign currency ones.

A more plausible cause for a CBDC to lose value relative to legacy currency, is, however, a design feature of some CBDCs. Both the European Central Bank and the Bank of England have – so far – declared their intention to limit the possible holdings of CBDCs. In  the case of the euro area, the limit being discussed is €3,000; whereas the Bank of England seems to be opting for a more generous £20,000.

This is a completely new situation. Until now, anyone who – for whatever reason – wishes to do so, has always had the legal possibility to go completely liquid – i.e., to sell all their assets and only hold cash. Now, at least some central banks are talking about restricting ‘safe’ holdings of money.

The reasons for such a policy are valid enough. They are there to ensure that, in a crisis, there is not an electronic bank run, where households and companies desert commercial banks and shift all their deposits to CBDCs. Were this to happen, banks would either have to replace lost deposits at once; or, more likely, shrink their balance sheets. Since that would mainly involve calling in loans, a major financial crisis would rapidly develop.

Nevertheless, setting a limit on how much CBDC anyone – households or companies – can hold, is a ground-breaking change with the current situation, where central bank money is available to anyone in unlimited amounts (unlimited relative to their total assets, needless to say). In addition, setting limits creates other problems, in particular if the limit is as low as the 3,000 suggested by the ECB. How would this apply to companies, where €3,000 is a laughably small sum? Even for individuals, it doesn’t take much or particularly large transactions to exceed €3,000 (selling a car, for instance, let alone a dwelling). According to Boris Vujčić, Governor of the Croatian National Bank, the CBDC accounts will function like waterfalls.[3] That is to say, the moment the amount exceeds the limit, it will automatically be transferred (presumably to a bank account). The CBDC accounts will be restricted to one account per person and they will not be remunerated. In essence, the CBDC will only function as a means of payment.[4] This is somewhat puzzling: The CBDC is supposed to be central bank money – that is its whole raison d’ȇtre. But, as described by Governor Vujčić, its usefulness will be limited, certainly compared with legacy currency. And finally, if there is a limit to anyone’s CBDC holdings, their usefulness as a tool of monetary policy becomes correspondingly limited.

(What about the opposite case, where the limits of CBDC holdings make them more attractive, rather than less? This is a point made in a working paper from the Bank of Japan, where the authors argue that “if the central bank imposes the limit on the amount of CBDCs, the exchange rates among CBDCs, cash and deposits would be destabilized since CBDCs may accompany scarcity premium.”[5] However, as has already been pointed out, if the divergence in value is due to an appreciating CBDC, this can easily be countered by the central bank.)

In addition, while far-fetched today, it is not beyond the realms of imagination to see a future where it may be illegal to use CBDCs for certain types of payments – e.g., alcohol, or tobacco (sugary foods even?), let alone drugs or prostitution. But even without such restrictions, it is possible to see a situation where a CBDC is perceived as less useful than legacy currency.

Finally, there is the possibility of concern about the safety and stability of the electronic infrastructure. In Sweden, an all but cash-less society, there was a run on ATMs at the news of the outbreak of the Russian invasion of Ukraine.

If you have two variants of the same currency, the fact that one of them is even slightly less useful than the other, could easily give rise to a divergence in value.

This may sound trivial, but it is not. If the CBDC were to diverge in value from legacy currency by depreciating, the central bank’s ability to conduct monetary policy would be severely impaired. Moreover, if there were no single cause for the divergence, restoring parity would be even more difficult.

From a longer perspective, this could also threaten the existence of central banks. If central banks, by introducing a CBDC, create financial problems and confusion; and if (as, in a number of countries) the use of cash is rapidly diminishing, the question why central bank money is needed in the first place, could well arise. Although CBDCs are at least partly intended to ensure that the non-bank private sector has access to central bank money, a discussion paper written by two employees of the Riksbank and one from the Bank of Canada[6], argues that central bank money is strictly speaking not necessary.

This is not an argument against the introduction of CBDCs. However, it is an argument in favour of moving carefully, and of considering how to deal with the possibility that the CBDC might be seen, not as a complement to legacy currency, but as a possible substitute, as different and might therefore diverge in value. It is perhaps also an argument for central banks to consider directing their energy towards improving already existing digital payments methods (e.g., the Swedish Swish system or the American Venmo), before moving into unchartered territory.

Gabriel Stein

(This article was previously published by Central Banking. It is based on chapter 5 in my forthcoming forthcoming PhD thesis One money – two values – Sweden’s era of parallel banknotes 1789-1803. It further develops the argument in that chapter.)


[1] This terminology is actually wrong. CBDCs already exist, primarily in the shape of commercial banks’ deposits with their central banks. But the term is already accepted and will be used here.

[2] See Ruchir Agarwal & Miles Kimball, Breaking Through the Zero Lower Bound, IMF Working Paper WP/15/224, October 2015; Katrin Assenmacher & Signe Krogstrup, Monetary Policy with Negative Interest Rates: De-Linking Cash from Digital Money, International Journal of Central Banking, Issue 67, March 2021

[3] Mentioned at a panel at the Euromoney Global Borrowers and Bond Investors Forum 2023, 20th June 2023

[4] Ibid.

[5] Yanagawa Noriyuki & Yanaoka Hiromi, Digital Innovation, Data Revolution and Central Bank Digital Currency, Bank of Japan Working Paper Series, No 19-E-2 February 2019, p 12

[6] Hanna Armelius, Carl Andreas Claussen and Scott Hendy, Is Central Bank Currency Fundamental to the Monetary System, Bank of Canada Staff Discussion Paper 2020-2