Central Bank Digital Currency… Does Not Have To Be Evil
Central Bank Digital Currencies (CBDC) can be used for both bank lending and for payments, without dis-intermediating the commercial banks.
In the process, CBDC can eliminate system risk due to bank failures, without compromising privacy, or the use of cash. They can also eliminate the ‘base cost’ of money, while providing Central Banks with a new way of managing inflation that is potentially more targeted.
CBDC’s also make it possible to protect deposits against inflation, while ensuring borrowers repay the real value of their loans, aiding economic efficiency… though such a feature is not a necessary condition of the switch to CBDC.
I’ve included a description of the accounting entries for each type of transaction to demonstrate that the system accords with current accounting practice.
Administratively, relatively little needs to change.
The biggest change is in the legal relationships between the Central Bank, banks and depositors, as well as formal recognition of CBDC as legal tender.
Money Creation, Allocation and Destruction
Most money is now created when commercial banks make loans. It’s as simple as making two entries in the books of the bank: debit the amount of the loan in the name of the borrower (which creates the debt owed by the borrower), and credit the same amount as a deposit in the same name (which the borrower can draw on to spend).
The new money allocated to the borrower is then spent into the economy to meet their needs, in the process driving new economic activity.
Unless taken in cash, payments are made by sending messages between banks to increase the account of the payee, while reducing the account of the payer by the same amount. Net settlements between banks and the Central Bank are done in similar fashion.
The money is destroyed as the loan is repaid and the entries in the books of the bank are reversed.
Essentially, it is ‘money out of thin air… and back again’.
Quite simply, the money itself costs virtually nothing to create. It is not like creating a ‘chair’ (for example), it just takes a bit of compute power and the press of a few keys to put numbers into the books of the Central Bank and the banks.
Of course, there are costs in managing the system, both at the Central Bank level and at the level of the banks, so that overall, the system is not cost-less.
CBDC would be no different, except the CBDC would be created ‘out of thin air’ by the Central Bank making loans to the banks. As now, the money would be destroyed as the loans are repaid. Again, the system itself would not be cost-less, though the creation of the money would be. How the system costs are paid is discussed later in the paper.
The CBDC would be legal tender, equivalent to cash. (This would require nothing but a simple change in the law).
Continuation of Cash
Cash could remain in the system, both in ‘paper’ form, and as eMoney on an electronic device, such as a phone.
The issue of cash or eMoney would not change the amount of money on issue. It would simply change its form from an entry in the books of the Central Bank, to either an amount of notes and coins, or a sum on your phone (or other device).
Cash is important as a ‘back up’ to keep the whole system functioning in the case of a widespread cyber attack on the payments system, or an attack or chaotic failure that brings down our power, or communications systems, or on the main internet connectors between capital cities and the rest of the world.
All people should (if possible) keep some cash in reserve. As well, banks should have stockpiles of cash at each branch for emergencies, with processes to continually dump changes to customer balances at each branch into a secure local server that can be accessed offline by branch personnel to enable the issue of cash on demand — if the system goes down. It would be clunky and slow, but better than the chaos that may ensue if all financial transactions are halted across all banks at the same time (shudder the thought). Even if the possibility is remote… it could happen. Risk management tells us that we need to take seriously risks that, though they only have a remote chance of occurring, they have existential consequences!
The Set Up
To shift from the current system to CBDC, the deposit accounts of each bank’s customers will need to be legally separated from the accounts of the bank.
Instead, all deposits would be recorded on a subsidiary ledger of the Central Bank (CB) managed by each bank for their own customers. Operationally, there would be no change to the payments system.
All existing deposits can be transferred in a single ‘round robin’ via a loan from the Central Bank to each bank. Each bank would then use the proceeds of the loan to pay out its depositors. It would do this by messaging the Central Bank to reduce its own account by designated amounts, while adding the amounts to newly created Central Bank accounts in the names of each of its customers.
These transactions would shift the deposits from the bank’s accounts, to the books of the Central Bank, using a newly established ‘subsidiary ledger’ of the Central Bank (managed by the bank).
Each bank would then have a liability to the Central Bank, in lieu of depositors.
While each depositor would then have a new Central Bank account (managed by their own bank). Operationally, customers would see zero change in how they conduct business.
Inter-bank loans could be similarly settled at the same time, though the details are beyond this paper.
Operating the Payments System While Maintaining Privacy
Under this proposal, banks would continue to operate the payments system.
When a customer made a payment, the payer’s bank would reduce the payer’s CB account by the amount of the payment, while messaging the payee’s bank to increase the payee’s CB account by the same amount. This could happen ‘instantly’.
Importantly, there would no longer be any ‘inter-bank’ balances, eliminating any potential flow-on effects in the event one bank fails.
When banks spend their own money, their CB account would go down, and the CB account of the payee would go up.
Money would never leave the Central Bank, unless cash is taken out.
Apart from each bank’s own transaction details (of spending on its own account), the Central Bank would only have the totals of each bank’s subsidiary deposit ledgers on its own books. It would have no customer or transaction details, ensuring the same degree of privacy, as now.
When a bank called on the Mint to provide cash, once the cash was delivered to the bank, the Mint would message the Central Bank to record the amount on the liability side of the Central Bank’s balance sheet as ‘cash on issue’. At the same time, the Central Bank would reduce the bank’s deposit account by the amount of the cash advance (which deposit also appears on the liability side of the Central Bank’s balance sheet).
The cash on issue goes up, as the bank’s deposit goes down, keeping the Central Bank’s book in balance.
The bank taking the cash would show it in their accounts as cash on hand (asset), while also reducing in its own accounts its deposit with the Central Bank (asset). Again, the cash goes up as the deposit goes down, keeping its books in balance.
When a bank depositor withdraws cash, the bank would cease to owe that amount to the Central Bank, as the cash would then be in circulation in the community. There are three steps to account for the cash issue.
First, the bank would show the transaction in its own accounts by reducing both its own cash on hand (asset), while increasing its own deposit with the Central Bank (asset) by the same amount, keeping the bank’s books in balance.
Secondly, it would reduce the depositor’s deposit on the Central Bank subsidiary ledger that is manages, while providing the depositor with ‘cash in hand’. The depositors deposit would go down, but their cash would go up by the same amount, keeping them in balance.
Thirdly, it would message the Central Bank to reduce (in the Central Bank’s books) the total of all deposits held on the bank’s subsidiary ledger (liability), and to add the amount back to the bank’s deposit account with the Central Bank (liability), also keeping the Central Bank’s books in balance.
By this process, the money is changed from an electronic entry in the books of the Central Bank (deposits), to ‘cash in hand’ held within the community. This has no impact on the amount of money ‘in circulation’, it simply changes its form from CBDC recorded in the books of the Central Bank, to cash in circulation.
In all cases, it remains as legal tender.
Lending and Interest Rates
Despite all deposit accounts sitting on the books of the Central Bank, the banks would retain their vital role in making local lending decisions, removing any need for the Central Bank to become involved in retail operations.
All advances would be made by each bank in CBDC, borrowed from the Central Bank.
When a bank agreed to advance a loan to a customer, it would request the Central Bank to credit the amount of the loan to the bank’s own deposit account at the Central Bank. In doing this, the Central Bank would record in its own books a debit, being ‘loan to the named bank’, matched by a credit being ‘deposit in the same name’. Keeping its own books in balance.
The bank would record the new CBDC deposit at the Central Bank in its own accounts as an asset (deposit at Central Bank), as well as the matching liability (loan from Central Bank). Again, keeping it’s books in balance.
The loan would then be extended to the borrower by a series of messages and ledger entries.
First, the bank would send a message to the Central Bank to reduce the bank’s deposit at the Central Bank by the amount of the loan.
Secondly, it would credit the borrower’s Central Bank deposit account with the same amount on the Central Bank’s subsidiary ledger (that it manages).
Thirdly, the bank would message the Central Bank to add the amount of the loan to the total of the bank’s subsidiary ledger account on the Central Bank’s books.
Again, the Central Bank would have no visibility into the specifics of the loan. It would simply see two entries in its books:
- The total of the bank’s subsidiary ledger balance (liability) would increase by the amount of the loan.
- The balance of the bank’s own deposit at the Central Bank (liability) would fall by the same amount.
These transactions would have zero net impact on either the Central Bank’s net worth, or the net worth of the bank. They would simply provide the money for the borrower to draw down.
Any payment made from the borrowers Central Bank deposit account would go straight to the payee’s Central Bank deposit account (perhaps managed by another bank). Or, it could be taken in cash (as described above).
In making the loan, the bank would show in its own accounts, a reduction in its own Central Bank deposit account (asset) equal to the advance, matched by a new asset being the loan to the borrower. Again keeping its books in balance.
There would be three major differences with the Existing System:
- Timing
Importantly, the loans from the Central Bank would be extended for the same period and repayable at the same rate as the loans made by each bank to each borrower. This would eliminate any ‘timing’ risk regarding repayment. (Currently, deposits are repayable ‘on demand’, while the loans are for ‘fixed terms’. This imbalance in timing is the source of greatest risk for the banking sector under the existing system)
2. The Cost of New Money Would Be Zero
The loans from the Central Bank to the bank would be interest free. This removes the ‘base cost’ of money from the system. (This would also be a major difference compared to the existing system, where bank loans must cover the cost of interest paid to depositors)
3. Payment of Central Bank Costs
Each bank could also pay a charge based on the number and size of the bank’s transactions to compensate the Central Bank for its costs in operating the system.
Impact: More Stable Interest Rates
In this situation, interest rates would be highly stable as they’d only have to pay for the bank’s operating costs, a provision for default, and its profit margin. The banks would have no ‘cost of funds’.
Managing the Money Supply
The Central Bank could still impact the money supply.
This could be done by adding an interest rate margin to any new loans to each bank. This cost would be passed on by the bank to new borrowers. This would deter some new borrowing, slowing the rate at which new money was spent into the economy, thereby reducing demand, and hence inflation pressure.
The extra margin could even be focused on a certain type of loan and in certain regions. To, say, deter borrowing for existing housing in an overheated market, while leaving borrowing for new homes and other spending unaffected.
If thought necessary, all loans could be impacted by simply applying the margin to existing borrowings as well.
This would be equivalent to increasing interest rates now to tackle inflation, though potentially better targeted.
Conversely, the Central Bank could encourage specific borrowing by offering an interest subsidy for new loans across specific sectors and/or regions, say, to promote new investment that is considered strategically important. Or, the subsidy could be across all new loans, or it could include even existing loans to free up cash to boost demand.
This would be equivalent to dropping interest rates now.
These decisions would be made in the context of the Central Bank’s twin mandate to maintain low inflation and low unemployment. Sector decisions would need to be taken in concert with the Government.
Targeted stimulus can also (or instead ) be provided though fiscal policy. Wider stimulus too may be provided through fiscal policy, or even a Universal Basic Income (both whole other topics!)
Money as Equity, Not Debt
Currently, all deposits and bank reserves appear on the liability side of the balance sheets of the banks and Central Bank, and are regarded as a ‘debt’ of the organizations that hold them.
Under the proposed system all Central Bank deposit accounts held by government, the banks, and people and organizations in the private sector, along with all ‘cash on issue’ would still sit on the liability side of the Central Bank’s balance sheet, but as ‘equity’.
There are two reasons why this is a better characterization of the reality.
First, the deposits and cash are not repayable by the Central Bank. Either payments go from one Central Bank account to another, or they are taken in cash, reducing the deposit with the Central Bank. Any re-deposit of cash simply does the reverse, reducing the cash on issue and increasing the deposit with the Central Bank. No claim can ever impact the Central Bank’s viability.
Secondly, the deposit accounts and cash on issue are not claims against the Central Bank, they are claims against the resources of society. Purely as an analogy, depositors are more like shareholders having a claim against the resources of the company (the country), than bondholders having a claim against the company itself (the Central Bank).
Prudential Oversight
Otherwise, banks will operate as now, with the Central Bank and APRA (in Australia) continuing to monitor each bank’s creditworthiness.
However, there is a major benefit.
With all bank debt being with the Central Bank, the CB can take a cyclical view of the system. This will allow it to support individual banks during any dip. That is, if a bank needs to provide extended support to its own customers (because of temporary personal difficulties), the Central Bank can provide the same extensions to the bank for those loans.
This would provide a major benefit, that would allow banks and their customers to ride out a temporary dip, without any risk to depositors
Another difference is the consequence of any permanent loss.
Unlike now (where the government can be called on to support a failing bank), any permanent loss on loans would be born by the shareholders of the bank, up to the limit of the bank’s capital.
Beyond that, any bank can be allowed to fail if they are no longer viable -without imperiling any other bank, as there would no longer be any need for inter-bank lending… as all accounts would be with the Central Bank.
The Central Bank can never become nonviable due to a bank failure as it simply creates the money in the first place. When its loans to any bank are repaid, the money is written back into the thin air from which it came.
Any lack of recovery of any loan simply means the money remains in permanent circulation, without impact on the Central Bank.
In the wind up of any bank, after all steps to recover the loans from defaulting borrowers had been taken, and after all shareholder funds had been exhausted, the net loss would be determined. This ‘net loss’ would be recorded in the Central Bank’s books by simply reducing the amount in its books labelled ‘loan to the (failed) bank’ to zero, and adding the same amount to a debit account labelled ‘money on permanent issue’. Again, keeping the Central Bank’s books in balance.
With all deposit accounts on the books of the Central Bank, no money on deposit would ever leave the Central Bank system, unless taken in cash.
This will remove both moral hazard and the threat of bank runs, as no deposit could ever be lost due to a bank failure (because the deposit would no longer be a liability of the bank).
Inflation Adjustments
It would also be possible to have the Central Bank credit every deposit with an amount equal to the daily inflation rate — tax free. This would ensure that the deposit retained its real value over time.
To compensate for this effective increase in the money supply over time, the Central Bank could also require that its outstanding loans to each bank be increased by the inflation rate. In this case, each bank would also adjust each of its outstanding loan principals on a daily basis, so that all borrowers had to repay the real value of their loan too. This would ensure that the banks were not out of pocket, while all borrowers had to repay the real value of their loan.
Given that outstanding loan balances roughly equate to total deposits, the two inflation adjustments would largely cancel each other, so far as the money flows into and out of the economy are concerned.
Importantly, these adjustments would result in lending and investment decisions that are more economically efficient. No one would need to put their money at risk, just to earn up to the inflation rate. Every investment should expect to earn above inflation to make it worthwhile taking the money off deposit.
By adjusting deposits to the inflation rate, it would mean that when you work or invest and create value measured by your deposit, you will be able to recover the same amount in real terms, whenever you spend… no matter how long you defer spending.
Conversely, if you borrow to consume value today, you’ll have to repay the real value of the loan over the period of the loan.
As mentioned at the start, the inflation adjustments would be a ‘nice to have’. They are not essential for the switch to CBDC.
Bank Trading and Investment
Ideally too, in return for banks being given access to interest-free new money for on-lending, we should restrict them from taking loans from the Central Bank for investment in their own name. Nor should they be allowed to lend to related parties.
Also, we should aim for banks to provide the first layer of debt, with the lowest risk. The actual risk level would be set by the bank, commensurate with the level supported by its shareholders.
That is, the primary purpose of banks would be to make relatively low-risk loans, and to operate the payments system.
Higher levels of risk would need to be taken on by people putting their own money at risk. In this case, investment opportunities would need to offer a return greater than inflation to induce investors to take their money out of their risk-free inflation adjusted Central Bank deposit account.
These provisions would aid market discipline and enhance market efficiency.
Full System Specification
This system is fully detailed in a paper co-authored with Biagio Bossone (who has previously co-authored papers on CBDC for the World Bank).
