Central bank digital currency: the end of monetary policy as we know it?

Marilyne Tolle.

Central banks (CBs) have long issued paper currency. The development of Bitcoin and other private digital currencies has provided them with the technological means to issue their own digital currency. But should they?

Addressing this question is part of the Bank’s Research Agenda. In this post I sketch out how a CB digital currency – call it CBcoin – might affect the monetary and banking systems – setting aside other important and complex systemic implications that range from prudential regulation and financial stability to technology, operational and financial conduct.

I argue that taken to its most extreme conclusion, CBcoin issuance could have far-reaching consequences for commercial and central banking – divorcing payments from private bank deposits and even putting an end to banks’ ability to create money. By redefining the architecture of payment systems, CBcoin could thus challenge fractional reserve banking and reshape the conduct of monetary policy.

The next big question for central banks?

Digital currency is no longer the preserve of cypherpunks and crypto-anarchists. Economists and central bankers alike have been pondering whether CBs should issue their own digital currency. Koning (2014) and Andolfatto (2015) have discussed the idea of ‘Fedcoin’, Ben Broadbent recently spoke on the possible technological underpinnings and consequences of a CB digital currency, and the People’s Bank of China has announced it is looking into the idea.

Show me the money

Cash is simply coins and notes – embodiments of ‘money.’ Because banknotes and coins circulate in the economy, they are also referred to as ‘currency’. Yet currency is only a very small part of money (see McLeay et al (2014)). Money mostly consists of electronic deposits: broad money consists of (currency and) households’ and firms’ deposits with commercial banks, while base or CB money consists of (currency and) commercial banks’ deposits with the CB (‘CB reserves’).

On the face of it, customer bank deposits and CB reserves are very similar. They are both current account balances. But there is a crucial difference. CB reserves are risk-free. Bank deposits are not, because banks engage in lending that incurs at least some risk. As Mervyn King (2010) remarked, the ‘pretence that risk-free deposits can be supported by risky assets is alchemy’. Commercial banks’ fallibility is the reason behind the existence of public deposit insurance and lending-of-last-resort by the CB – an attempt to enforce one-for-one convertibility between bank deposits and CB money.

What might happen if the CB were to issue digital currency?

Now assume a CB issued CBcoin, a digital currency with one-for-one convertibility with paper currency and CB reserves. The issuance of CBcoin would simply create a third CB liability, risk-free and irredeemable.

The first step would be to decide whether, and at what interest rate, CBcoin might be remunerated. CB reserves are the means by which most CBs today implement monetary policy, by setting the interest rate paid on the reserves (or via the rate on repo transactions).

If CBcoin were remunerated at the same rate as CB reserves, it would be interchangeable with reserves. And if the CB chose to replace cash with CBcoin, it could then charge a negative interest rate on deposits to bypass the dreaded zero lower bound, as considered by Kimball (2013) and Haldane (2015). In this scenario, the overall quantity of CB money would stay the same, only the composition of the CB liabilities would change.

But even if the CB didn’t use the price or quantity of CBcoin as an additional monetary policy instrument, CBcoin issuance could have much wider ramifications, as a by-product of its impact on the payment system.

An overhaul of transactions settlement?

CB reserves currently play a central role in payment systems. If two parties need to settle a transaction but hold deposits at different banks, the payment requires a transfer of funds between the two banks. Banks net out such transfers and settle the residual amount using CB reserves as the medium of exchange.  This makes CB money the ultimate settlement asset (see Rule (2015)). While some intermediated electronic payments such as the UK’s Faster Payments Service are fast, traditional systems can be slow, taking up to three business days to settle a transaction (see Kroeger and Sarkar (2016) and Yermack (2015)).

If households and firms were given access to CBcoin accounts at the CB, banks’ dominant role as providers of payment services would be called into question. As a risk-free, interest-bearing asset, CBcoin would be preferable to bank deposits (and even paper currency, presuming anonymity concerns were addressed), encouraging households and firms to convert their bank deposits into CBcoin deposits. The appeal of CBcoin vis-à-vis deposits would likely depend on the relative interest rate payable.

In effect, retail payments (and securities transactions) would no longer have to be mediated by banks, as the funds would be transferred directly from one party’s CBcoin account to another’s. A disintermediated payment system could gradually replace the current centralised system and its associated credit and liquidity risks (see BIS (2003)). The main benefit to CBcoin account holders would be access to cheap and fast peer-to-peer transactions.

An end to traditional banking?

Commercial banks currently have the power to create money. When a bank makes a loan, it simultaneously creates a deposit, adding to broad money. So, by extending credit, banks not only create their own funding (deposits), they also control the level of broad money in the economy (see McLeay et al (2014)). Banks hold a fraction of the loans they extend as CB reserves, so as to back a fraction of their deposit liabilities with CB reserves – a setup known as fractional reserve banking. This fractional backing of deposits means that if all households suddenly wished to convert their deposits into hard currency, banks would not have enough reserves to repay them, so would either need to sell off their loan books in exchange for currency or utilise the CB’s lender-of-last-resort facilities.

If households and firms converted their bank deposits into CBcoin, commercial banks’ deposit-funded model would come under pressure. Broadly speaking, there are two possible delimiting scenarios. In the first, banks would compete with CBcoin by offering higher interest rates on their customer deposits. How much higher would of course be an empirical matter. By raising banks’ funding costs – other things equal – this could dent bank profitability and lead to tighter credit conditions. But banks would continue to issue loans and create broad money.  In a recent paper, Barrdear and Kumhof use a DSGE model that accommodates fractional reserve banking to study the macroeconomic consequences of CB digital currency issuance.

Another scenario would see a large-scale shift of customer deposits into CBcoin, forcing banks to sell off their loan books. Bank deposits could still exist but as saving instruments, no longer used to make payments. Banks could still originate loans, provided they lent money actually invested by customers, say, in non-insured investment accounts that couldn’t be used as a medium of exchange. Banks would operate like mutual funds, losing their power to create money and becoming pure intermediaries of loanable funds, as described in economic textbooks.

Under this scenario, the contraction of broad money (bank deposits), and the attendant emergence of ‘private-sector base money’ made of CBcoin would mark the demise of fractional reserve banking (see Sams (2015)). The conversion of bank deposits into CBcoin deposits at the CB would amount to 100% reserve backing for deposits. This could usher in a system similar to the Chicago Plan, a set of monetary reforms proposed by Irving Fisher during the Great Depression and recently revisited by Benes and Kumhof (2012). The Plan’s call for the separation of the credit and money-creating functions of private banks would be addressed – with 100% reserve backing, banks could no longer create their own funding – deposits – by lending.  Similar “narrow banking” proposals have emerged since the financial crisis, such as that of Kay (2009), Kotlikoff’s Limited-Purpose Banking (2012) or the Vollgeld initiative (2015), recently rejected by the Swiss government.

A new framework for monetary policy?

The conflation of broad and base money, and the separation of credit and money, would allow the CB to control the money supply directly and independently of credit creation, calling for a reassessment of monetary policy along two dimensions. First, the prospect of direct control of the money supply might alter the relative merits of using interest rates or the money supply as the main policy instrument. If so, this newfound CB power could reopen the debate between advocates of rules versus discretion in the conduct of monetary policy. For instance, the signers of the Chicago Plan, in particular Milton Friedman, envisioned a constant money growth rule rather than the discretion over interest rates that has prevailed since CB independence in the 1990s.

Watch this space!

Mapping out the implications of CB digital currency issuance is a very complex endeavor. It is helpful as a first pass to sketch out partial scenarios, as I have done in this post for banking and monetary policy, but the devil lies in the detail. Research is ongoing so watch this space!


Marilyne Tolle works in the Bank’s MPC Unit.

If you want to get in touch, please email us at bankunderground@bankofengland.co.uk. You are also welcome to leave a comment below. Comments are moderated and will not appear until they have been approved.

Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.


Filed under Banking, Currency, Monetary Policy

12 responses to “Central bank digital currency: the end of monetary policy as we know it?

  1. Marcel Polk

    Interesting article. However, it is not clear to me why these implications are a result of CBcoin. If the CB would just allow individuals to hold deposits with the CB, the consequences would be exactly the same in my view. This could be done today, but CBs have normally considered this not its task (the practical implications were, of course certainly before the internet area, very cumbersome). Today the CB could consider such move and let individuals open a current/deposit account with the CB (but again unless this is considered CBcoin it could be done today under the current rules). Am I missing something?

  2. Farhad Moshiri

    ” Banks could still originate loans, provided they lent money actually invested by customers, say, in non-insured investment accounts that couldn’t be used as a medium of exchange. Banks would operate like mutual funds, losing their power to create money and becoming pure intermediaries of loanable funds, as described in economic textbooks.”

    Who would create money in the system then? The pro regarding current monetary system lies in the ability of the banking system not to impose rationning on the quantity of money.

  3. David Acorn

    I think you need to understand what gives our FIAT currency value in the first instance. The answer to that in the UK system is, taxation. The government requires us to pay our taxes with Pounds Sterling. We have to get some Pounds from somebody to pay the taxes that we can’t dodge.

    There is only one place you can get Pounds Sterling from and that is the government’s Treasury; it is the currency ISSUER, (not the BoE) it has a bottomless pit of the stuff that it will never run out of, because it is FIAT. There is no bill, presented to it in Pounds Sterling, that the Treasury can’t pay. Likewise, there is nothing that is for sale in Pounds Sterling, that the Treasury can’t afford to buy.

    When the Treasury pays my pension, it “digitally” transfers a series of numbers from a Treasury spreadsheet to my banks spreadsheet into the cell with my account number on it. My bank now has a liability that is not backed by an asset. The Treasury effectively puts an equal amount to my pension, into my banks current account at the BoE, where it is called a “reserve”; my banks balance sheet balances again. If I send my pension to another bank, the “reserve will follow it to that other bank in clearing.

    The last eight years have proved that one of the two tools of “monetary policy” in the form of QE, (that is the BoE swapping Gilts back into the reserves – Treasury previous spending – that originally bought them) does not increase aggregate spending in the economy, it just distorts asset prices.

    It is time that interest base rate were permanently put to zero; The BoE was put back into the Treasury Department as the national clearer and regulator. You may be aware that the WGA accounts consolidates the Treasury and the BoE as one and the same. Start using a dynamic Fiscal Policy to regulate the economy and maximise the use of labour and capital resources.

  4. Mathias Mond

    Thank you for the heads up. I am no expert in banking or finance but rather an engineer/IT person. As such, I would like to hear from you how a transition from the current system to any such new system could be managed, with a possible dramatic reduction of the role and business of current commercial banks. The dynamics of such a transition could become very interesting, especially in reference to transitioning current assets and liabilities in light of the duration mismatch while keeping commercial banks in business, and cross-border aspects if not everyone is moving at the same time. I apologize if this is a stupid question.

    Second, I would not rely on our Internet infrastructure to be reliably available at all times (it certainly isn’t right now in some corners of the world). The effects you are sketching out such as direct control of money supply would only arise if alternate systems (fractional reserve banking, cash) are phased out or at least radically diminished. What are the potential fallbacks to make such a system resilient?

  5. Jim Tuller

    If I ask why CBcoin might be any different from cash, I am led to consider a valuable service performed by banks which is overlooked by this analysis. It is said that because of CBcoin, retail payments no longer would have to be mediated by banks. Yet is this not true of cash? It works this way in underground economies which can function very nicely without recourse to the central bank. It was true in pioneer communities, which had no banks.

    Security is an essential service provided by banks. Let us not forget that digital currencies have been hacked and stolen repeatedly. Without banks, who will secure and guarantee our ownership of CBcoins? Given the current state of digital security, you might be better advised to keep your cash in the proverbial mattress.

    The late Utah Phillips, an American folk singer, would tell a story of the hobos, itinerant workers who (illegally) rode freight trains around the United States as they travelled in search of seasonal work. When they would get paid, they would deposit funds with a local restaurant which would maintain a punch card tally for them. These workers were vulnerable to robbery. As Mr. Phillips explained, “Even if you didn’t much care for the restaurant, if you got rolled, you still could eat.” Even an undistinguished restaurant was better than no bank at all.

  6. Kim B

    It would be interesting to see this in conjunction with peer to peer lending.

  7. Jacky Mallett

    The most striking thing about current Economic Theory is its almost complete failure to investigate and correctly analyse the fractional reserve banking system in all its many current and historical variants. Instead we have macro-economic models that are rudimentary at best, and even in their latest incantations (Kumhof et al) are still based on overly simplified cartoons of the financial system.

    While this is harsh, if there is wide spread agreement on anything at the moment it is the failure of macro-economic theory.

    In that context, I would suggest that what you are proposing is to convert
    the existing banking system so that it resembles the one being described in these eminently falsifiable models, in the hope that the economy will then behave as predicted by these models. This is about as sensible as it would have been to allow 15th century European astronomers to modify the orbit of the planets into epicycles in order to fit their heliocentric theories. Especially since unlike epicyles (which did in fact manage a fairly good fit to observations) these models have to be extensively “calibrated” (over fitting to the curve on steroids) to have any resemblance at all to observed data.

    More prosaically – there is no particular magic about fractional reserve banking, it is simply a form of statistical multiplexing. There is no reason why fractional reserve banking couldn’t be performed using CBCoin, and in reality, it would be quite hard to prevent.

    It might be better to study and properly understand the existing financial system, before getting too carried away with replacing it. The 20th century historical precedents for the ‘We had to destroy the economy in order to save it route’ are rather less than compelling.

  8. It has been surprising me for some time that governments and central banks don’t see that by providing convenient electronic cash system, they could take back control of money supply, which is currently left largely in the hands in the commercial banks. Well done for realizing, you hit the nail on the head, this would constitute real banking reform and could avert the next fractional reserve based credit bubble and ensuing crash.

    I fear the next mistake will be to assume that the blockchain is the right technical solution.

  9. John Mesrobian Esq.

    I prefer Cash and Gold vs Cryptocurrency/Bitcoin/Digital currency. I am not alone, the vast majority of the world still prefers and uses Cash vs electronic money.

    Then we have the security of this Cryptocurrency/Digital currency, from hackers.

    Also, using Digital Currency, it would be easier for Confiscation by Gov’ts in times of troubles and etc.

    But, Digital Currency does have a place and a role but, real Cash and Gold still are KING and King for sometimes to come.

  10. Sorry @John Mesrobian Esq. but only about 3% of the fiat currencies are cash. All the rest is electronic. The difference between electronic money and cryptos is that account for electronic money is hidden and crypto account is open; anyone can see its accounts. Which do you think is more honest?