What’s wrong with cash?

It is widely believed that in times of uncertainty people will turn to cash. It may or may not be a general rule, but the past year has shown it to be true in Australia during the COVID-19 pandemic. The value of banknotes on issue haa risen over 17% compared with an average increase of only 5% per year during the previous decade,¹

Is cash being ‘hoarded’? It probably depends on your definition of ‘hoarding’. Most of the increase was in $50 and $100 notes. The authors above also note that the cash wasn’t used for purchases since the value and number of cash transactions fell during the year.

It seems safe to conclude then that a significant amount of cash has been kept under the mattress in anticipation of …… In anticipation of what? Nobody knows, but it seems that people feel that ‘cash is king’ in times of uncertainty.

Does it matter? The central banks of the world think so. At a time when central banks are trying to expand bank deposits, citizens are reducing them by hoarding cash. The value of banknotes in circulation, or under the mattress, reached almost $100 billion by February 2021,²

To put the amount in perspective, recall that the RBA’s initial commitment to ‘quantitative easing’, aka ‘printing money’, was $100 billion. The amount is, therefore, significant, and it is one of the reasons why central banks around the world are beginning to show an interest in ‘Central Bank Digital Currency’, CBDC for short.

What is CBDC?

There are almost as many forms of CBDC as there are central banks. For our purposes, it is enough to classify them as ‘retail’ CBDC and ‘wholesale’ CBDC. ‘Retail’ CBDC is digital currency issued by a central bank that is intended for general use, perhaps replacing paper banknotes as the major form of legal tender.

‘Wholesale’ CBDC is intended for use primarily in the clearing and settlement system. The RBA has partnered with some private firms to investigate the use of blockchains and wholesale CBDC in the settlement system: see the RBA website for the Media Release of 2 November 2020. ³

The use of private blockchains is trendy but totally uninteresting. The blockchain would necessarily be a ‘private’ blockchain,⁴ and a leading expert has noted that such blockchains are merely an inefficient form of:

distributed add-only data structures with a limited number of entities authorised to add data. Such systems have been studied for over 60 years and are blockchains in name only. The only reason to use one is to ride on the back of the blockchain hype.(Schneier 2019)

Central banks have, of course, used electronic money in the form of accounts for many years. It is hard to see the attraction of replacing Exchange Settlement Accounts with some other form of ‘digital currency’.

Which leaves us with the much more interesting ‘retail’ CBDC. Important discussion papers have been issued by, among others, the Bank for International Settlements,(BIS 2020) the International Monetary Fund,⁵ the Bank of Canada⁶ and the Bank of England.⁷ Our own RBA has a number of Research Discussion Papers and Bulletin articles relevant to the subject.⁸

The closest to a definition of a retail CBDC is found in the discussion paper issued by the BIS.⁹ According to that discussion paper, CBDC:

• is a central bank liability;
• is denominated in an existing unit of account;
• serves as a medium of exchange; and
• serves as a store of value.

The IMF considers ‘retail’ CBDC as ‘a widely accessible digital form of fiat money that could be legal tender.’¹⁰

Who wants CBDC?

The casual observer could be forgiven for thinking that CBDC is a solution in search of a problem. The Bank of Canada assures us that it is not planning to issue CBDC anytime soon, but will consider it if ‘certain scenarios materialize or appear as if they are likely to’. These ‘scenarios’ are:(Jiang 2020)

• the use of bank notes were to continue to decline to a point where Canadians no longer had the option of using them for a wide range of transactions; or
• one or more alternative digital currencies—likely issued by private sector entities—were to become widely used as an alternative to the Canadian dollar as a method of payment, store of value and unit of account.

The second ‘scenario’ is the important one. Without naming it, the concern is with the Facebook sponsored stablecoin known originally as the ‘Libra’ but changed to the ‘Diem’ in late 2020. The Diem will be the subject of a future note.¹¹

The Bank of Canada and other central banks are a little too modest about the ‘advantages’ of CBDC. Other, more forthcoming, proponents claim multiple advantages. For example, (Bordo and Levin 2018) claims that CBDC provides:

• an almost costless medium of exchange;
• secure store of value if interest is paid on CBDC;
• gradual replacement of paper money; and
• true price stability: the real value of CBDC would remain stable over time in cpi terms.

The first ‘advantage’ probably reflects a US-centric view of the world since the US payments system lags behind that of Australia and other parts of the world. The BIS paper notes that a good payments system might make the CBDC less attractive.¹² ‘Good’ was interpreted to mean ‘fast (even instant) and efficient’. Australia’s New Payments Platform fulfils this role nicely.

The third ‘advantage’ probably should not be too widely advertised. It seems unlikely to be popular with those that like cash. Acceptance, if even possible, might depend on privacy provided by the system.

The ability to pay interest on CBDC holdings and, less well advertised, to impose penalties in the form of the more palatably named ‘negative interest’ is probably the real reason that central banks are currently so interested in CBDC. As will be seen below, CBDC provide a new channel for implementing monetary policy.

What about privacy?

Privacy in digital currency systems, whether CBDC or private digital currencies, is a complex topic. Part of the problem is the number of participants in a typical payment transaction. Aside from the payer and payee, there will typically be several banks, a central bank and various, perhaps numerous, government agencies who will wish to be able to probe the details of individual transactions.

Bitcoin is often held out as the ultimate payment system for providing privacy. The privacy offered is often attributed to its use of the blockchain. This is wrong from a technical point of view. A blockchain could be completely public, identifying participants by name or by some other public identifying token. Bitcoin was designed to be anonymous by using public key cryptography to hide identities.

The same cryptographic technology could be used to make normal bank accounts anonymous. The obstacle is not technology, it is public policy that demands access to movements of money.

It is obvious then, that the degree of privacy offered will depend on the detailed design of a CBDC system. (Darbha and Aroroa 2020) provides a detailed description of the range of privacy options available for different designs, showing in the process that CBDC design decisions are far from trivial.

One thing is clear: CBDC design will allow certain agencies to penetrate transactions to prevent money laundering and terrorist financing. The major discussion papers mentioned above all include money laundering and terrorist financing considerations as an essential part of the design criteria.

What about monetary policy?

The ability to set the cash rate is a fundamental tool used by central banks. The RBA has, as part of its response to the pandemic, set the cash rate as close to zero as can be imagined. What about moving to a negative setting? If the bank were to contemplate a negative setting, how far could it go.

Central bankers talk about the ‘lower bound’ problem. It is easy to explain: a ‘negative cash rate’ is nothing more or less than a penalty on deposits. Although there are costs associated with large cash holdings, there comes a point where the penalty for holding deposits is more than the cost of holding cash. Rational citizens will withdraw deposits and hold cash.

According to the analysis of CBDC proponents, that sets a mark below which the cash rate cannot fall, a ‘lower bound’. They argue that the problem would go away if it were possible to pay interest on cash holdings. Of course, what they really mean is if it were possible to dip into the cash holdings which they describe as paying ‘negative interest’.¹³.

Not everyone is convinced that CBDC is the only way to solve the lower bound problem even if it is a problem.¹⁴ The Bank for International Settlements also concluded that existing tools could achieve the same results but without introducing unforeseen risks that might arise from CBDC.¹⁵.

Does CBDC have a future?

The literature on CBDC is already vast and conflicting. About the only thing that all commentators can agree upon is that acceptance by the public will depend on the design and distribution model of any CBDC scheme. The RBA is actively considering these problems. ¹⁶

CBDC may have a future, but it is unlikely to be a near future in Australia. We already have an advanced, efficient payment system which eliminates one of the major claimed advantages of CBDC. The government of the day is unlikely to allow any CBDC that prevents active intervention to detect money laundering or terrorist financing activity, so privacy will inevitably be a concern. Although not discussed here, security of any CBDC system is obviously of the utmost concern.

At least for the foreseeable future, ‘cash’ will mean the folding stuff that you hide under your mattress. All $100 billion of it.