Digital Currency Design Principles

Joseph Huber


Central bankers as well as monetary reformers are discussing the introduction of central bank issued digital currency (DC). DC is intended to coexist with bankmoney (sight deposits), both monies circulating side-by-side and in competition with each other, similar to the familiar coexistence of bankmoney with traditional cash. Money users would opt to maintain their current bankmoney account, or to use a DC device or digital currency account as the new alternatives, or to use bankmoney and DC in parallel, depending on the prevailing conditions and individual preferences.

First design studies of DC were put forward by Barrdear and Kumhof of the Bank of England, the Swedish Riksbank and the Basel Bank for International Settlements, and were also presented at an early stage by monetary reformers and other economists [1].  The number of central banks and international monetary institutions who have expressed an interest in DC has been steadily growing [2].

If a central bank is the monetary state authority of a respective nation or community of nations, the means of payment issued by that central bank represent sovereign money. It is a defining feature of sovereign money to have the status of unrestricted legal tender. From this angle, the question arises of whether DC can be seen as a step towards a sovereign money system, over time replacing the present bankmoney regime. The answer to that question depends on the extent to which a number of design principles as discussed below are going to be implemented.

Money-on-account or cryptocoin?

Initially, DC was imagined in the form of cryptocoin based on distributed ledgers and blockchain technology. However, the new technology is still in its infancy. The transfer of cryptocoins is not yet sufficiently fast, and is much too energy-intensive and thus expensive and environmentally damaging. Cryptocoin trading platforms are vulnerable to hacker attacks, and legal questions of liability and identifiability are not entirely settled.

In comparison, tried and tested ways of managing account balances and payments from and to accounts are well suited for implementing DC. In the foreseeable future, DC is thus likely to be implemented in the form of account balances (money-on-account). At the outset, this may not involve immediate individual access by nonbanks to the central-bank payment system; instead, access to DC by way of mobile apps and e-cash cards, and later also through an additional infrastructure of DC accounts affiliated to the central-bank payment infrastructure [3]. This does not exclude the future application of cryptographic technology to managing DC.

Potential advantages

Among the reasons for seriously considering the introduction of DC are the disappearance of cash and a related far-reaching loss of monetary control. Bankmoney has marginalised central-bank money in all advanced countries, presently at the ratio of 80–95% bankmoney to 20–5% cash in euro countries’ M1 (stock of money in public circulation) [4]. The actual share of cash is even smaller, as more than half of it is kept in safes or used as a parallel currency abroad. On top of this there are money market fund shares as a new money surrogate largely based on bankmoney, amounting to 2.4 times M1 in the U.S. and a third of M1 in Europe (UK included) [5]. An unspoken worry related to the long-term decline of cash is that a central bank with no central-bank money in public circulation might seem somewhat redundant, sort of King Lackland.

If the shift towards bankmoney could be reversed, however, resulting in a growing proportion of central-bank money in M1, the transmission lever of conventional monetary policy could be expected to become correspondingly more effective again [6]. Other advantages of DC include the complete safety of DC in contrast to bankmoney, which is inherently unsafe and must be backed by a number of auxiliary constructions of uncertain reliability (such as, for example, equity ratios, deposit insurance and government bail for bankmoney). The safer the existing stock of money is, the more this will contribute to overall financial and economic stability.

As far as DC is concerned, there is no need to prop up banks in a crisis in order to save the nation’s money and to maintain payment transactions. In terms of comfort and costs, the handling of DC is likely to be equal to that for bankmoney. Regarding the costs to the banks for funding DC, the situation is comparable to that of cash. A further advantage related to this, particularly for the public purse, is increased seigniorage in proportion to the stock of DC.

Setting course

One might assume that as soon as DC is available there will be strong migration from bankmoney to DC. However, DC may not in fact be a fast-selling proposition, and its advantages may not materialise automatically. For example, one would not expect firms and people to feel urged to switch accounts:

under conditions of business-as-usual when there is no sense of heightened uncertainty,
if central banks and governments maintain far-reaching state guarantees for bankmoney, and
if banks pay some deposit interest on bankmoney, while none or much less is paid on DC.

Under such conditions, it remains unclear whether a significant shift from bankmoney to DC would occur at all.   

What then are the conditions and design principles that can be expected to support a gradual switchover from bankmoney to DC, so that central-bank money would over time again be the dominant and system-defining means of payment?   

No restrictions on access to and relative quantities of DC

The first principle is to secure countrywide access to DC devices and currency accounts according to customer demand. In most proposals put forward thus far, DC is rightly intended to be a universal means of payment. What does the universality of DC come down to, if not to the wider principle not to restrict the use of DC, in terms of actor groups or quantity of DC available?

In contrast to this, access to DC is reserved only for financial institutions in one model variant [7]. In an earlier concept paper, the quantity of DC was restricted to 30% of GDP [8]. In the Swedish concept, the use of e-kronas is not expressly limited, but only DC devices would be available at the beginning of the process (mobile apps, DC cards), and these are subject to the legal ceilings on cash payments in Sweden, currently at a maximum equivalent to about 250 euros (285 dollars) for each transaction [9]. Why would this make sense? Qui bono? Such limits and restrictions clearly contradict the claim of DC to be a universal means of payment. Should the non-financial public even be excluded from using DC, the whole project would in fact be pointless.

Merging DC and interbank reserves into one circuit

The next design principle is the merging of DC and interbank reserves into a single circuit. Thus far, the English and Swedish proposals have kept reserves and DC apart from one another. This is another arbitrary and implausible design feature. ‘Reserves’ and ‘digital currency’ both refer to the same kind of central-bank money-on-account, and there is no difference regarding the digital nature and the monetary quality of the central-bank money involved.

A desirable design principle is therefore to treat excess reserves like DC in general. This means:

merging the banks’ excess reserves and their DC, and
maintaining free exchange between the banks’ and nonbanks’ DC, thereby creating a single DC circuit, involving banks as much as nonbanks.

This does not imply a blurring of the difference between a pure DC transaction account and a bank’s central-bank refinancing account, nor does it impair monetary policy [10].

Full convertibility between bankmoney and DC

A subsequent principle is full convertibility between bankmoney and DC. Bankmoney must be freely convertible into DC, and vice versa. Technically this poses no problem, as can be seen in the example of bank-mediated payments between the central-bank transaction accounts of state bodies and bank giro accounts of nonbanks.

Convertibility of bankmoney into traditional cash was, and essentially still is, a prerequisite for the acceptance of bankmoney and its parity with central-bank money [11]. This will also apply to the conversion of bankmoney into DC, particularly as both monies offer the ease of cashless payment.

Central bank guarantee of converting bankmoney into DC, particularly in a bank run

Convertibility of bankmoney must all the more be ensured in a bank run situation. In actual fact, warranted convertibility of bankmoney is the definite answer to the bank run problem. This is to say that in a bank run situation, central banks should stabilise banking and finance not by trying to stop the bank run, but: by supporting the conversion of bankmoney into DC.

To this end, central banks would have to implement QE by granting special (presumably unsecured) credit to banks specifically for the conversion of bankmoney into DC. The measure itself would effectively prevent banks from going bust and would actually help forestall a bank run situation altogether.

Gradually reducing and ultimately removing state warranties of bankmoney

There is also the question of retaining or withdrawing state warranties for bankmoney. As long as these government guarantees are maintained, combined with basically unrestricted pro-active bankmoney creation, one cannot seriously expect the introduction of DC to eventually lead to a situation in which sovereign money would again be system-defining. Therefrom, another design principle is to reduce and finally remove the state guarantees of bankmoney. The bigger the share of DC has become, the more the state guarantees of bankmoney can be withdrawn.

Gradual increase in the use of currency accounts by public bodies

Some of the payment transactions of public bodies are carried out today via transaction accounts with the central bank, and many others via bankmoney accounts. It is among the absurdities of the present bankmoney regime that state bodies demand to be paid in private bankmoney rather than in the sovereign currency of the state’s central bank.

Public bodies should therefore be obliged to conduct transactions via currency accounts. However, the state’s acceptance of bankmoney is a key pillar in the state’s warranty of bankmoney. Should this pillar be removed too fast, with public expenditure at 35–55% of GDP depending on the country, bankmoney would be undermined in a way similar to a run on bankmoney. Nevertheless, public bodies can begin to use currency accounts in addition to bank giro accounts, steadily increasing their use of DC.

Central-bank credit to banks not the only channel for issuance of DC

The Swedish and English concepts of DC continue the practice of issuing central-bank money by way of credit to banks against collateral. The Swedish model involves converting bankmoney into e-krona (which presupposes the banks to have created bankmoney as well as the central bank to sell or lend e-kronas to the banks). The English model issues DC by way of central-bank purchases of sovereign bonds from financial institutions (which presupposes the financial institutions to have acquired the larger part these bonds with bankmoney rather than reserves). Either way, it is not the central bank but the banks themselves who in the first instance decide whether and how much money is created, while the central bank continues to accommodate the facts the banks have created beforehand.

However, DC also can and ought to be issued in a direct way. This would include measures such as helicopter money. It might also include revising Art. 123 (1) and (2) TFEU (also known as Lisbon Treaty). In its present form, this Article is overtly inconsistent in that its first clause prohibits direct monetary financing of government expenditure, while the second clause indirectly permits monetary financing of sovereign bonds.

Central bank deposit interest on DC equal to deposit interest on bankmoney

In the concept variants by Barrdear/Kumhof and Kumhof/Noone, DC is interest-bearing. In the Swedish concept, by contrast, the e-krona does not yield interest. Why would DC be interest-bearing? Interest is paid on credit and debt positions, or say, on promissory items. DC, however, is not a promissory position. It is fully existing sovereign fiat money in its own right, base money that does not need coverage by another kind of money or collateral.

Why then do some scholars argue the case for interest-bearing DC? One reason is to create a peg on which to hang negative interest (next paragraph). Another reason is ‘to clear the market’ [12]. The macroeconomic idea of market equilibrium, however, is empirically hard to substantiate. By contrast, it is quite obvious what deposit interest on DC actually can do: it complements the deposit interest on bankmoney that banks are likely to pay so as to prevent deposits from draining away.

If there were to be deposit interest on bankmoney, but none on DC, this would significantly contribute to an undesirable effect of pro-cyclical fluctuation: conversion of bankmoney into safe DC in times of heightened uncertainty, and back to interest-bearing bankmoney in times of business-as-usual. In this regard, paying deposit interest on DC can be a neutralising measure if the interest rate paid on DC is equal to the interest rate on bankmoney. This will create a level playing field and counteract the undesirable pro-cyclical shifting.

Ruling out ‘negative interest’

The question of negative interest is not specifically related to DC, but is also relevant to DC [13]. The problem here is that abstract arithmetic does not necessarily fit the real world. For example, ‘real interest’ is commonly defined as the actual interest rate minus the inflation rate. The result may be positive or negative. Either way, however, it is a matter of combining two different classes of operands. This certainly makes sense when considering the actual-versus-nominal purchasing power of various kinds of income (earnings, interest, transfers), but it does not make the inflation rate an interest rate.

Seen from another perspective, an individual can have a greater or lesser income, or no income at all, but no negative income; less than nothing does not exist. Breaking through below the ‘lower bound’ is possible in the world of numbers, but not in the real world. What in fact can happen, for example, is a loss of purchasing power and wealth, or even incurring debts. Hence, as has been said often enough, negative interest is an unnatural concept. You pay interest to someone who has lent you money, but you do not agree to pay interest to someone who has borrowed from you. Similarly, it would be nice to go shopping and to have the shopkeeper pay you the purchase. Apparently, this would be turning the real world upside down.

Negative interest is an inappropriately expanded and hence distorted measure of conventional interest rate policy, in a desperate attempt to regain the latter’s effectiveness which has largely been lost in the present bankmoney regime. What actually happens when ‘negative interest’ is imposed is as follows:

Negative interest on bankmoney reduces the liabilities of banks to their customers and results in higher balances of a bank’s profit account. This is tantamount to an illegal private tax on deposit money to the benefit of the banks. At the same time, the stock of bankmoney is reduced.

The removal of liabilities from the banks’ balance sheets, that is, deletion of bankmoney, would certainly contribute to reducing the existing overhang of money that is the inheritance of the bankmoney regime. However, a reduction in this way is wrongly targeted, as it hits the mass purchasing power in the form of the income and savings of the middle classes.

Similarly, using present accountancy rules, negative interest on DC would reduce the central bank’s liabilities and thus far the available stock of DC. The resulting profit in terms of the central bank’s equity would be paid out annually to the treasury, adding to the public purse. Negative interest would then indeed be a tax on holdings of DC; without therefore becoming more sensible economically.

It needs to be seen that negative interest misses its aim of stimulating expenditure that would result in demand-induced growth, on the grounds of trying to evade negative interest by spending the money as soon as possible. It remains open to question under what conditions this kind of ‘economic policy by monetary policy’ might be reasonable. Independently, most people react in different ways. Negative interest, rather than spurring faster or additional expenditure, is also likely to trigger compensatory spending cuts. If money is confiscated from people, they do not hurry to spend what is left, but try to make up for what has been taken away (except under conditions of runaway inflation). Negative interest is a technocratic folly born from unworldly model economics.

Negative interest is sometimes interpreted as a surcharge on top of the service fees for managing accounts and payments. This too is hardly plausible and instead appears to be a form of extortionate pricing.

Imposing negative ‘interest’ is actually neither about interest nor fees; it is about the overt expropriation of money if imposed on bankmoney, and is an unwise tax if imposed on reserves and DC. As an instrument of monetary and economic policy, negative interest is counter-productive and unjust, perhaps even unlawful, and should thus generally be ruled out, also in connection with DC.

Concluding remarks

On reading central bank statements on DC, one is left with an ambivalent impression, as if central bankers were running ahead of their convictions. Problems taken from thin air are put forward against DC, for example an allegedly impaired ability of banks to lend and to invest, or the bank run problem – which in fact is a problem; not, however, of DC, but the general threat inherent in bankmoney and fractional reserve banking.

The modern world has been living for 150–300 years with the conflicting situation arising from the coexistence of sovereign money and bankmoney, at first in the form of private banknotes alongside sovereign coin, and subsequently in the form of bank deposit money alongside central-bank money (legal-tender banknotes and reserves). The equally conflicting situation formed by the coexistence of bankmoney and DC will not be very different from this.

Nonetheless, the introduction of DC in parallel with bankmoney is a step forward, and to a certain extent offers the advantages mentioned above. The problems inherent in the present near-complete rule of bankmoney are still much bigger than problems relating to DC might be.


Joseph Huber

The author is professor emeritus of economic sociology, Martin Luther University, Halle, and co-founder of the monetary reform initiative Monetative in Berlin.


  1. Barrdear/Kumhof (BoE) 2016 3–18, Kumhof/Noone (BoE) 2018 4–22, 35–37, Sveriges Riksbank 2017, 2018, Ingves (Sveriges Riksbank) 2018, Dyson/Meaning (BoE) 2018, BIS 2015, 2018, Bech/Garratt (BIS) 2017, Niepelt (Swiss National Bank) 2015, 2018. Pioneering inputs from monetary reformers were made by Dyson/Hodgson 2016, Wortmann 2016, 2017a+b, Yamaguchi/Yamaguchi 2016, Huber 2017 188–190, 2018 [2014]. Other economists supporting DC include Bordo/Levin 2017, Bordo 2018, Eichengreen 2017, Roubini 2018.
  2. Recent studies include IMF 2018 and Central Bank of Iceland 2018.
  3. Cf. Ingves 2018 4 [9], Sveriges Riksbank 2018 pp.15.
  4. ECB, Economic Bulletins, Tables 5.1.
  5. Huber 2017, 111.
  6. Cf.  Dyson/Meaning 2018.
  7. Kumhof/Noone 2018 pp.18, where three model variants are discussed: (1) access for financial institutions (FI) only, (2) economy-wide access for everyone, and (3) FI only combined with narrow banking based on DC, which in fact comes down to continued reserve banking.
  8. Kumhof/Noone 2018 pp.18, Barrdear/Kumhof 2016 3, 50.
  9. Sveriges Riksbank 2017 21.
  10. The IMF study on DC, too, concludes that ‘CBDC is unlikely to affect monetary policy transmission’ (IMF 2018 4, 25).
  11. Ingves 2018 2 [9], Bjerg 2017, 2018.
  12. Kumhof/Noone 2018 pp.8.
  13. Among those who see DC as a suitable vehicle for imposing negative interest are, for example, Bordo/Levin 2017 3, Bordo 2018 3. The IMF study on DC also states that DC ‘would eliminate the effective lower bound on interest rate policy’, even if the central banks surveyed in the study declared negative interest not to be a reason for introducing DC (IMF 2018 4, 29).


Barrdear, John / Kumhof, Michael. 2016. The macroeconomics of central bank issued digital currencies, Bank of England, Staff Research Paper No. 605, July 2016.

Bech, Morten / Garratt, Rodney. 2017. Central bank cryptocurrencies, Basel Bank for International Settlements, Quarterly Review, September 2017, 55–70.

BIS. 2015. Digital currencies, prep. by the BIS Committee on Payments and Market Infrastructures, Basel: Bank for International Settlements. November 2015.

BIS. 2018. Central bank digital currencies, prep. by the BIS Committee on Payments and Market Infrastructures, Basel: Bank for International Settlements, March 2018.

Bjerg, Ole. 2017. Designing New Money – The Policy Trilemma of Central Bank Digital Currency, Copenhagen Business School Working Paper, June 2017.

Bjerg, Ole. 2018. Breaking the gilt standard. The problem of parity in Kumhof and Noone’s design principles for Central Bank Digital Currencies, Copenhagen Business School Working Paper, August 2018.

Bordo, Michael D. / Levin, Andrew T. 2017. Central bank digital currency and the future of monetary policy, NBER Working Paper Series, no. 23711, Aug 2017.

Bordo, Michael D. 2018. Central Bank Digital Currency. The Future Direction for Monetary Policy? Shadow Open Market Committee, E21 Manhattan Institute, March 9, 2018

Central Bank of Iceland. 2018. Rafkróna? Central bank digital currency, Special Publication No.12, September 2018.

Dyson, Ben / Hodgson, Graham. 2016. Accounting for sovereign money. Why state-issued money is not ‘debt’, London: Positive Money, January 2016.

Dyson, Ben / Meaning, Jack. 2018. Would a Central Bank Digital Currency disrupt monetary policy? Bank Underground, 30 May 2018.  

Eichengreen, Barry. 2017. Central bank-issued digital currency is the future, not cryptocurrency, CNBC The Fintech Effect, 30 Oct 2017.

Huber, Joseph. 2017. Sovereign Money – Beyond reserve banking, London: Palgrave Macmillan.

Huber, Joseph. 2018 [2014]. Digital Currency. Retaining or overcoming the bankmoney regime? Design principles that make the difference.

IMF. 2018. Casting Light on Central Bank Digital Currency, IMF Staff Discussion Note, written by T. Mancini-Griffoli et al., Washington D.C.: International Monetary Fund.

Ingves, Stefan. 2018.  The e-krona and the payments of the future, Speech by the Governor of the Swedish Riksbank, Stockholm.

Jakab, Zoltan / Kumhof, Michael. 2018. Banks are not intermediaries of loanable funds – facts, theory and evidence, Bank of England, Staff Working Paper No. 761, October 2018.

Kumhof, Michael / Noone, Clare. 2018. Central bank digital currencies – design principles and balance sheet implications, Staff Working Paper No. 725, May 2018, London: Bank of England.

Niepelt, Dirk. 2015. Reserves for everyone – towards a new monetary regime?, VOX Policy Portal, 21 Jan 2015,

Niepelt, Dirk. 2018. Reserves for all? CBDC, deposits, and their (non-)equivalence , Centre for Economic Policy Research, Discussion Paper 13065, London, July 2018.

Roubini, Nouriel. 2018. Why central bank digital currencies will destroy Bitcoin, The Guardian, 19 Nov 2018,

Sveriges Riksbank. 2017. The Riksbank’s E-Krona Project, Report 1, Stockholm, Sep 2017.

Sveriges Riksbank. 2018. The Riksbank’s E-Krona Project, Report 2, Stockholm, Oct 2018.

Wortmann, Edgar. 2016. A proposal for radical monetary reform, Amsterdam: Ons Geld.

Wortmann, Edgar. 2017a. Deleveraging without a crunch, Ons Geld Working Paper, Utrecht,

Wortmann, Edgar. 2017b. The virtual euro, Ons Geld Working Paper, Utrecht,

Yamaguchi, Kaoru/Yamaguchi, Yokei. 2016. Peer-to-Peer Public Money System, Japan Futures Research Center, Working Paper No. 02-2016, Nov 2016. 

Leave a Reply

Your email address will not be published. Required fields are marked *