The case for a central-bank currency register

Accounting for sovereign money on banks' and central banks' balance sheets 

Schreibtisch Buchhaltung.jpg

This paper deals with the question of how a transition from bankmoney to sovereign money can best be accounted for by central banks and banks. For a start, readers familiar with the matter may read the section on 'core elements recapped' at the end of the paper. Independently, and prior to addressing accounting more specifically, the properties of a pure or partial sovereign money system that are of particular relevance to accounting will be outlined.

Properties of a sovereign money system in contrast to the present
bankmoney regime

Sovereign money reform replaces bankmoney with central-bank money. To date, sovereign money exists in the form of coins and notes. Such solid cash comes from a currency area's central bank and enters into general public circulation via the banks. Sovereign money also exists in the form of bank balances with the central bank, called reserves. Reserves are central-bank money-on-account, 'account money' for short. A smaller amount of reserves represents so-called excess reserves for making payments only within the interbank circuit, while the rest constitutes the minimum reserve requirement – if minimum reserves have not been abolished altogether (such as in the UK), because they represent a mostly idle stock of central-bank money that is of little monetary use except for generating, in normal circumstances, some interest revenue for a central bank (paid by the banks).

As a result of sovereign money reform, customers' account money in the form of demand deposits in a bank is converted into central-bank money, that is, sovereign account money for public circulation among nonbanks, thus in fact replacing the bankmoney in the public circuit among nonbanks. Cryptocurrencies, should they achieve the status of a ubiquitous means of payment and be issued by central banks, would also be sovereign money, competing with bankmoney and replacing it fully or partially.

In countries where the central bank is a monetary state authority, the terms sovereign money and central-bank money are interchangeable in most contexts. Presumably this also applies to the European Central Bank (ECB) as an intergovernmental body. The extent to which this also applies to each national central bank in the Eurosystem remains open to question.

The transition from bankmoney to sovereign money can basically be achieved in two ways:
-    either by means of a complete D-day transition, immediately bringing about a pure sovereign money system with a five-to-ten-year period of phasing out old bankmoney liabilities,
-    or gradually, whereby bankmoney on the basis of fractional reserve banking and sovereign money beyond reserve banking exist side by side, similar to the present   coexistence of bankmoney and solid cash.

Traditional cash, however, appears to be heading towards its end, while, if introduced, sovereign account money, and perhaps also sovereign cryptocurrency, could be expected to spread rapidly. The idea of offering sovereign account money in parallel with bankmoney once was briefly mentioned by J. Tobin. He called it 'deposited currency', but did not expand further on it.[1]  

Sovereign cryptocurrency would constitute another money circuit of its own, similar to cash, in that the currency is transmitted from payer to payee without the monetary intermediation of a bank as a trusted third party – directly from digital wallet to wallet, as cash circulates from hand to hand. A growing number of central banks are currently communicating their interest in generating a variety of 'digital currency' of their own. So far, however, few details of such plans have been published.[2] The often indistinct usage of terms like e-cash, electronic money, money-on-account (account money), digital currency and cryptocurrency (based on distributed ledgers and blockchain technology) creates some confusion. A few central banks, or individual central bankers, express reservations about central bank digital currency. In both the supporters and the critics, however, it is not always clear whether they refer to sovereign account money or sovereign cryptocurrency.  

How would sovereign account money and sovereign cryptocurrency be brought into circulation? To the extent that the bankmoney regime continues to exist, central-bank reserves are created and issued by central-bank credit to banks. Traditional cash and maybe cryptocurrency can be issued by central-bank credit to banks further on but would also be issued by transfer of genuine seigniorage from the central bank to the public purse.

In a pure sovereign money system, there would no longer be reserve banking or bankmoney. All money in whichever form would come from the central bank and have the status of legal tender.
The major part of it would be issued long-term by way of genuine seigniorage, which is the gain from creating new money and spending rather than lending it into circulation.
The minor part of additions to the stock of sovereign money, as far as it is needed for monetary policy operations, would be emitted short-term by way of central-bank credit to banks or for other open market operations or for taking in foreign exchange against domestic currency.
The criteria for the creation of new money by a central bank and the issuance of the money either by genuine seigniorage or by central-bank credit to banks do not fall within the scope of this paper.[3]

Introducing sovereign account money for use by the wider public, thus potentially for all money users, will contribute to shifting the dividing lines and overlaps between banks, shadow banks and other financial institutions. In a pure sovereign money system, banks are no longer monetary institutions participating in an exclusive interbank circuit on the basis of reserves, with access to central-bank refinancing and the central-bank payment system. Banks can no longer re-finance at only a small fraction of the bankmoney they have created. Instead, they have to finance all of their lending and investment activities in full through their current proceeds and revenues, repayment of loans and other released assets, taking up money from customers and on the open market, issuance of bonds and other types of securitized debt, perhaps by increases in equity capital, and, in the last instance, borrowing from the central bank within the limits set by the current monetary policy. As is already apparent today, money and payment services will be carried out by more service providers than just banks. Equally, the banks are already facing strong competition from other financial institutions and fintechs in the lending and investment business as well as in wealth management. Banks and non-monetary financial institutions will increasingly resemble each other.

Within the system of fractional reserve banking, the meaning of 'bank' is unambiguous. A bank is a monetary financial institution, in contrast to non-monetary institutions (funds and trusts of any kind, securitization vehicles, wealth managers and insurance companies). Banks create bankmoney; non-monetary institutions are not able to do so, any more than other nonbanks in general. In a pure sovereign money system, such differences do not exist anymore. Money creation no longer forms part of a bank's business.

As for central banks, they still act as banks in a sovereign money system, for example in managing the national stock of foreign currency, lending to banks short-term on a limited scale, or carrying out open-market policy transactions. At the same time, a central bank has always been a particular kind of bank, as an originally private but privileged 'bank of the state', and as 'bank of the banks', over time acting as the carrier of a nation's monetary sovereignty and monetary policy. The commercial business of central banks has receded into the background, while their institutionalization as monetary authorities has increasingly come to the fore.       

This applies in spite of the fact that today it is primarily the banks that act as quasi-sovereign creators of bankmoney, issued in one act with extending bank credit on a fractional base of reserves. The central banks are backing the inherently unstable private rule of bankmoney by re-actively and routinely providing the reserves that the banks still need. In times of crisis, the central banks act as the banks' almost unconditional lenders of last resort by providing 'whatever it takes' to keep failing banks afloat, in cooperation with the government as the bankmoney guarantor of last instance.

This is the case for the simple and compelling reason that, in today's state-backed rule of private bankmoney, a nation's money and bank credit are firmly welded. The money supply today consists of 85–95 per cent pro-actively created bankmoney, systemically overriding everything else and especially pre-determining the central banks' re-active fractional provision of reserves and residual cash. Thus, saving the nation's money in a crisis and ensuring the money supply means saving the banks and ensuring their survival. In the final result, the needs and interests of the banking industry have to be accommodated at the expense of anybody else's needs and interests.

Back to top > 

Money and credit

To overcome the deficiencies of the bankmoney regime, a sovereign money reform resumes a basic Currency School tenet, which is the separation of money and credit, more precisely speaking the separation of money creation by the central bank from the uses of money in banking and the financial and real economy beyond. The banking sector no longer has the quasi neofeudal privilege of creating itself the money on which it operates. In particular, sovereign money reform segregates a bank's own means for proprietary lending and investment from the customers' money (as far as customers do not explicitly lend money to a bank).

The currently predominant Banking School doctrines, regardless of whether they are held in a neoclassical or a post-Keynesian context, allege the identity of money and credit. This is combined with labelling bank credit as endogenous or inside money, while central-bank credit is labelled as exogenous or outside money. Viewed in this way, the separation of money and credit seems to be illogical and even absurd, for present-day money is credit-borne bankmoney in the first place.

Another view holds that credit has decoupled from money. According to Schularick/Taylor, former varieties of money-based financial capitalism ('age of money') have been succeeded by credit-based financial capitalism ('age of credit').[4] Put pointedly, credit has been replacing money. Such event, however, is impossible. Bank credit cannot decouple from bankmoney, because the bankmoney automatically accompanies the extension of bank credit. In the same way, central-bank credit to banks invariably creates reserves or results in the issue of cash.    

Credit is not money, and paying out the amount of credit in kind or a transfer of securities cannot generally replace money as a regular means of payment in a useful and efficient way. According to presently fashionable teachings, however, everything that has monetary value is or can be used as money, particularly securities of any kind, that is, claims on money in lieu of the money itself. The means of payment and promises to pay are thus utterly confused – a confusion inherent in bankmoney, which is itself a promise to pay rather than money proper.

Of course it is true that anything of monetary value, anything with a price attributable to it, can be used as a means of payment if the parties involved agree on the deal. This, however, does not mean that everything of monetary value is money. Only the special tokens that are most widely, generally and normally used as a means of payment, that is, liquid assets in ubiquitous use for the settlement of any kind of debt, can be rated as money. Money is neither credit nor debt but a medium for paying out or paying back credit and eliminating credit-related or purchase-related debt.

By crediting a customer account, a bank creates transferable deposits that are used like money in cashless payments: habitual bankmoney indeed. But the bankmoney and the credit contract are nevertheless two different things, different legal matters fulfilling different functions in the economy. The bankmoney is an account-registered claim on legal tender or on sovereign money, respectively. It circulates among many subsequent money users, while the original credit contractor, the debtor, remains the same person all the time. It is not the credit (the loan) that acts as money, but the deposit, the bankmoney.

The large volumes of secondary credit among nonbanks are based on bankmoney. MMF-shares too are created by putting bankmoney into an MMF that in turn invests the money in short-term securities.[5] The use of MMF-shares as a deposit-like means of payment is widespread but of limited scope in that they are most often used in financial market transactions only. Beyond MMF-shares, payment by transfer of equity shares or other securities is limited to special cases, for example company mergers and acquisitions. Trade bills are used today less often than was formerly the case, and in most cases they are not used as a means of payment but as collateral for taking up money. Securitization of credit claims (asset-backed securities) also serves the procurement of liquid money. In the modern world, there is no credit extension without an immediate or final flow of money. The thesis of credit decoupling from money is overly misleading as is the false assumption of their identity.

Fashionable expert opinions obliged to Banking School teaching have created a Babylonian confusion of language regarding the notions of money, credit, bank and related further categories. Allegedly it cannot even be pinned down clearly what money is and what it is not – which is astounding in a science that claims to be bound to non-interpretive exactness.

What has escaped the attention, however, and to which the Schularick/Taylor decoupling thesis actually applies in a sense, if at all, is the fact that bankmoney has to a large extent become detached from central-bank money; more precisely speaking, the fact that the primary pro-active, for the most part overshooting and crisis-prone creation of bankmoney fully determines the re-active and fractional re-financing of banks in central-bank money. As one result, the effectiveness of conventional instruments of monetary quantity and interest-rate policy has greatly deteriorated. An opposite situation would actually make more sense: a monetary system under effective control of a state's independent central bank providing a flexibly re-adjustable stock of 'high-powered' safe and stable sovereign money, leaving the lending and investment business to the banking and financial sector. That it is what sovereign money reform is about. 

Back to top > 

Previous approaches to central-bank accounting for sovereign money. Money as a liability  

What is a suitable approach to accounting for sovereign money replacing bankmoney, or accounting for their coexistence? It is certainly desirable to keep the established accounting standards as far as possible. On the other hand, the relevant rules need to be adequate for the form of money and its issuance. The answer to that question is hardly to be found in the existing IFRS/IASB and GAAP/FASB standards.[6]

In considering accounting, present-day monetary reformers first simply followed the established conventions. The initial idea was to keep the familiar rules while being well aware of the altered meaning of a respective item. For example, the transfer of genuine seigniorage to the public purse was imagined to be accounted for as a credit claim of the central bank on the treasury, with the particularity of representing open-ended credit free of interest, in fact the oxymoron of debt-free debt. The counter-party to credit and a creditor is debt and a debtor. Pretending that in the case of genuine seigniorage the government is incurring debt just formally and not in reality is not really convincing.

One related rationale was that a central bank delivers its profit to the treasury, much of it coming from interest-borne seigniorage from lending domestic and foreign currency. Why should this not apply to genuine seigniorage too? Genuine seigniorage, however, does not derive from interest and does not show up as income in the profit and loss account. Genuine seigniorage is not generated by credit but by the sovereign unilateral act of creating fiat money. The rationalisation of debt-free debt and the idea of genuine seigniorage as a profit when there are no prior business events generating that profit represent a highly problematic overstretching of the meaning of credit and a distortion of the meaning of liability or debt.[7]  

Back to top > 

What no longer holds true in accounting for money and credit

Private banknotes from the 17–19thcenturies were issued through bank credit to customers. Since the 19th century the same applies to the issuing of central-bank notes by central-bank credit to banks. The credit money is created by a simultaneous pairwise entry into the books as a claim on the banks and a liability to the banks. The practice had a point at the time, because notes were not pure fiat money yet but promissory notes, a kind of bearer debenture, liable to conversion into silver coin on the demand of the bearer. Since the beginning of the note monopoly of national central banks in the 19th century and the underlying gold standard of the time, paper money was generally based on the principle of covering the official paper money with a corresponding national gold hoard.

Present-day central-bank money is no longer covered by other monetary items but is genuine pure fiat money in its own right. Bankmoney, in turn, is hardly covered by more than just a very small fraction of it available in central-bank money. For 100 euros of bankmoney, the banking sector in the Eurosystem needs to have available on statistical average about 1.4 euros in coins and notes for the ATMs, 0.1–0.5 euros in excess reserves for interbank payments and 1 euro that is the largely idle minimum reserve, together constituting 2.5–3 per cent of the entire stock of liquid bankmoney in the public money supply M1.[8] 

Bank credit, as well as central-bank credit, is normally secured by collateral in the form of bonds, other securities, or property. These act as credit default insurance so that, in a given case, the creditor can take advantage of these assets. This, however, does not mean to cover or back up the money, the latter being pure fiat money that need not be covered monetarily. The validity of money largely depends on the government establishing a means of payment as legal tender, or accepting a money surrogate such as bankmoney for its own uses. The value of the money, its purchasing power, is covered by what money can buy, primarily by the current economic output (= income). This also serves as value anchorage for financial and real assets, even though the dependence of asset prices on the current economic output is of a more indirect and complex nature.

In spite of the above situation, central-bank credit is still accounted for as if it were a promise to pay something other than itself. With regard to the banks, the situation is slightly different but similar. Bankmoney (bank deposits) is a bank promise to convert the account balance into cash or transfer it to customers at other banks by transferring an equal amount of central-bank reserves to the recipient's bank. Cash and reserves, however, are available only to a small fraction of the bankmoney. If too many customers want to withdraw too much cash or transfer too much bankmoney at once, the banks will be illiquid and even insolvent (if not 'saved' once more by central-bank and government intervention).

Cash, moreover, is no longer constitutive for the money system, even though its use in small-change-transactions is still widespread in many countries. Cash no longer has a monetary origin of itself. Rather, it is withdrawn from a bank account. The extension of credit, and thus the creation of bankmoney, precedes its conversion into cash. As banks have been stripped of the ability to create banknotes themselves since the 19th century, they have to refinance cash at 100%. But this no longer carries much weight, as the share of cash is now down to about 5–15% of the money supply M1. That is why the banks can manage with a cash reserve of just 1.4% of their bankmoney liabilities.

As regards reserves, in the sense of central-bank account money, the customers will not receive them anyway. The 'high-powered' reserves are reserved for bank use only, while the customers obtain the surrogate, that is, bankmoney as a promissory entry on their bank account. The bankmoney is transferred from one bank account to another within and between the banks. If in-between banks, the accompanying interbank transfer of reserves is carried out nowadays by a central-bank RTGS payment system but perhaps also by simple interbank clearing of claims and liabilities without any immediate flow of central-bank money. Consequently, bank credit is not really paid out to the customer, although this is what the relevant credit law requires. The laws differ depending on the country, but what they require is much the same: the credited amount of money has to be segregated from the creditor's assets and entered into the debtor's possession.[9] Instead, the money is entered into another bank's possession.

The fact that bankmoney does not have to be paid out in reserves – although this is what the law requires – enables the practice of fractional reserve banking. The main reason is that the interbank circulation of reserves is many times faster, i.e. much more frequent, than the customers' use of their bankmoney. By and large, the use of bankmoney in public circulation follows a monthly rhythm, while the ongoing use of liquid reserves in the interbank circulation is a question of seconds, especially in today's very large banks.

What is more, a bank's own payments are not segregated from the customers' payments. Both are managed via one and the same central-bank account of a bank. A transfer of customer account balances within a bank does not involve central-bank money anyway but is performed by reposting the bank liabilities from one customer account to another. Within today's megabanks this is a significant contributory factor.  

Overall, the practice of accounting for bank credit, or bankmoney, respectively, by a pairwise entry of a credit asset and a bankmoney liability has long since become unreal and inconsistent to a degree that some critics consider the practice of fractional reserve banking as a kind of legalised fraud.[10] In any case it is not easy to see why a money system would be functional when it makes banks and customers each other's creditor and debtor at the same time.

Regarding central banks, the liability practice of accounting for credit-borne central-bank money has become inconsistent in much the same way, most recently since the end of the gold dollar in 1971, the definite historical end of gold standards altogether. Why would a central bank, after disbursement of a credit to a bank, still owe something to that bank? Conversely, why would a bank still have some fictitious claim to have the money converted into some other monetary item? (except for the material warranty of substituting unworn banknotes for worn ones). The sovereign creator of fiat money, today in the form of solid cash and reserves on account, lending such sovereign money to banks, has a credit claim on the debtor banks, not a 'liability' to them, while it is the banks that have the liability to the creditor central bank.

Back to top > 

Accounting for sovereign money according to the coin or equity approach

Taking into consideration the above assessment, the liability approach to accounting for sovereign money was succeeded by a much better idea, which is to proceed with sovereign money and genuine seigniorage in a way that is analogous to the accounting for coins to date, treating sovereign money as an asset and equity, an addition to the national monetary endowment, rather than treating it as a liability. That coin or equity approach – with differences in the details, but in accordance with the basic idea – has been suggested by Jackson/Dyson and Benes/Kumhof as well as, independently, by Gudehus and both authors by the name of Th. Mayer.[11]

Coins are booked into a central bank's balance sheet as an asset. The coins are then sold to the banks at face value against reserves. When sold, the coins leave the balance sheet. The proceeds do not add to the central bank's profits, as the coins are bought from the treasury at par. Most treasuries traditionally still hold the prerogative of coinage. They bear the production costs of the coins and rake in the gain, the seigniorage, as the difference between the coins' face value and their cost of production.    

Unlike coins, notes and reserves enter a central bank's balance sheet in connection with credit extensions to banks as a liability to those banks. This does not create genuine seigniorage but results in interest gains from the banks' interest payments to the central bank. That gain is nowadays also called seigniorage, interest-borne seigniorage, even though it concerns a banking margin profit rather than seigniorage from money creation strictly speaking. The amounts of central-bank money issued by way of credit extension remain as a liability on the central bank's balance sheet until they are closed out on redemption. 

The idea now was to enter newly issued central-bank money (reserves, notes) – that is, new sovereign money – onto the balance sheet as an asset like coins, no longer as a liability, which basically results in an addition to the equity account. On the disbursement of newly created money to the public purse, the money is closed out on both sides of the balance sheet, resulting in a reduction of the assets as well as the equity. The stock of central-bank money in circulation can thus no longer be seen from the balance sheet but would be registered and disclosed in separate statistics.      

The issuance of sovereign money by way of central-bank credit to banks was initially not considered in the coin or equity approach. However, at least some part of newly created money needs to be treated in this way, because the money supply needs to be re-adjusted flexibly in the short-term, complementing and correcting the long-term issuance of genuine seigniorage. When this necessity of monetary policy is taken into consideration, this will continue to result in a degree of interest-borne seigniorage. This would not include, however, the conventional liability accounting. Instead, there would be an asset swap, whereby the credit claims on banks are swapped for the liquid money when the latter is transferred to a bank account.

The coin or equity approach to accounting for sovereign money still has small blemishes. This does not only refer to the fact that the amount of central-bank money in circulation can no longer be seen from the central bank's books. More importantly, rather than overstretching the notion of credit, as the liability approach does, the coin or equity approach overstretches the notion of equity. In connection with this, no author has explained by which account the money would add to the equity. This can hardly be imagined as a capital injection into a central bank's equity account analogous to that process in normal companies. Money creation is certainly different from a stock of liable capital. Even less imaginable is the use of a sub-account in the profit and loss statement. Sovereign money and related genuine seigniorage do not result from banking profits but are brought into existence by a sovereign unilateral act of fiat money creation. That act certainly has manifold preconditions, but previous business events subject to double-entry bookkeeping are not related to it. In consequence, a different new class of equity account, an equity account for money creation, would have to be introduced for central banks only – which would in fact be fairly straightforward.

Back to top >

Separating a currency register from the banking business of central banks

In view of the pros and cons of the approaches discussed so far, an earlier approach appears to be worth considering anew. It is not far from the coin and equity approach, rather an extension of it, but may be more appropriate in some of the details. The approach dates back to Ricardo's Plan for the Establishment of a National Bank from 1824. Ricardo was the most prominent representative of the Currency School. According to his concept, the Bank of England was divided into an issue department, responsible for the note issue, and a banking department, concerned with the Bank's banking business, that is, foreign exchange, refinancing the banks, and account management and payment services for important banks and a number of state institutions, at the time also including important trading houses and wealthy families.        

The separation of the note issue from the Bank's banking business exists to the present day. The arrangement, however, soon turned out to have little relevance even then, the main reason being that the separation of money and credit was not pursued with the necessary consistency. Country banks were allowed to continue with the private note issue for a long time. The gold standard was temporarily loosened or even suspended for a while (immediately resulting in the overshooting of credit creation and the railroad crises of the time). Furthermore, only the note issue was subject to the Bank Charter Acts, while bankmoney creation was left unregulated, despite the fact that cashless payment by the clearing of claims and liabilities via bank giro accounts then already played a role in England, becoming ever more important over time. Finally, the issue department emitted its notes to the banking department by way of credit and double-entry bookkeeping against private bills or government bonds.[12]  The issue department thus continued to act like a commercial bank rather than a sovereign money creator. Central banks were still privileged private banks rather than the national monetary authorities they have become today.  

Insufficient implementation of a principle does not invalidate the principle as such. Greater success of a new attempt depends on identifying a consistent way of keeping money creation apart from a central bank's banking activities. The approach suggested in this place intends to put the money creation into a separate ledger off the central bank's balance sheet. That ledger for money creation might be called the currency register. It does not constitute a balance sheet on its own, nor would it enter into the central bank's balance sheet based on double-entry bookkeeping. Money is fed into such a register by sovereign acts of fiat money creation, which do not presuppose a prior business event. As explained in the sections above, double-entry bookkeeping cannot sensibly be applied to sovereign money creation.      

The currency register is a separate ledger of the central bank, but it does not require a separate body to run it. Decisions regarding money creation and the currency register need to rest with a central bank's top management, for example the ECB's governing council and executive board, or the board of governors and the federal open market committee of the U.S. Federal Reserve, or the governors and board of directors of the Bank of England. In terms of monetary policy, a central bank's currency register and its banking activities connect to each other. The currency register would regularly be disclosed just as the major items on the central-bank balance sheet. 

The currency register is a daily ledger, an accountancy journal. It registers the current creation and accumulated stock of money together with the basic way of issuing the money; nothing more and nothing less. The taking in of foreign exchange against domestic currency continues to be managed on the central bank's balance sheet. The same applies to the management of the stock of foreign exchange.

The currency register might record the relevant operations in the following accounts:
-    the currency account and its sub-accounts, recording the different means of payment
-    the issue account and its sub-accounts, recording the basic channel by which the means of payment are released into circulation. 

The currency account serves the creation of sovereign means of payment as well as possibly their deletion or exchange. The currency account is sub-divided into accounts for coins, notes, today's reserves or, more generally speaking, sovereign account money, and possible future sovereign cryptocurrency.    

The money comes into existence on its registration in the currency account, in that a particular amount is simultaneously booked in as a particular means of payment in one of the currency sub-accounts as well as emitted either as genuine seigniorage, be it for government expenditure or citizens' dividend, or as a deposit onto the central bank's balance sheet, adding to its currency reserves (see Table 1).

Table 1   The account structure of the currency register's journal  

Table 1 Currency register accounts.png

Even though the currency register is not based on double-entry bookkeeping, each entry would consist of the time and amount of money simultaneously ascribed to (a) one of the four or five currency accounts and (b) one of the two or more issue accounts. The currency and issue accounts in the table can be sub-divided further if necessary.

From today's perspective, new central-bank money (sovereign money) is likely to enter circulation in the form of reserves. These may soon also serve as sovereign account money in public circulation.

Metal coin and paper notes remain the traditional alternative to bankmoney to the degree that these are still in demand or as long as their supply is maintained by banks and politics. Prior to their emission, they are not recorded in the currency register. Only when the pieces of metal and paper are registered as emitted currency do they begin to exist as money.[13] Even though solid cash is unrestricted legal tender, ever more public and private agencies, including the revenue office and banks, restrict the use of cash or even refuse it.

Sovereign cryptocurrency might in future become an adequate replacement for solid cash. Sovereign account money circulating among nonbanks fulfils the same function. The more widespread a cryptocurrency's use as a general means of payment becomes, the more its supposed 'anonymity' will turn out to be a myth. Sovereign cryptocurrency would be emitted by the currency register in the same way as other forms of sovereign money.

All forms of sovereign money can be exchanged for one another, also including private bankmoney as long as it exists. 

Together with an amount of money being recorded in the currency register, that amount is assigned to a channel of issuance in the issue account. The issue account has at least two sub-accounts:  
-    Genuine seigniorage, that is, the transfer of the money to the public purse,   irrespective of whether it will be used for government expenditure or a citizens' dividend;  
-    Central bank liquidity, that is, depositing the money as an addition to the currency reserves on the central bank's balance sheet. The money might be earmarked for the purchase of foreign exchange or for the short-term re-adjustment of the domestic money supply by directly lending short-term to banks or engaging in wider open-market operations.  

The transfer of seigniorage to the treasury is not about monetary financing of government expenditure in the sense of Article 123 (1) TFEU. It is about seigniorage indeed, which a national government is entitled to, of an amount deemed to be appropriate according to the central bank's independent monetary policy. The Parliament and Cabinet ought not to be allowed to demand money, issue directives, or interfere otherwise in this respect.[14] How much money is being created, and how much of it will be issued long-term as genuine seigniorage or by way of short-term credit to banks, shall be exclusively the central bank's business in fulfilling its legal mission – which, however, would have to be specified much better than is the case today.

With regard to current seigniorage, it should be understood that the pursuit of capacity-oriented monetary policies does not involve unheard-of amounts of money. Depending on the level of government expenditure and economic growth, the seigniorage involved could be expected to amount to about 1–4 per cent of total public expenditure annually.[15] That is no small change but far too little to contribute to public households on a large scale or to supplant the regular tax funding of public households. Seigniorage does not absolve the government from the necessity of good housekeeping.

Seigniorage by itself does not determine the purposes for which the money is spent. In the spirit of separation of powers between monetary and budgetary-fiscal responsibilities, the uses of seigniorage should not generally be fixed. Of course, however, there are different national and political preferences. Quite a few authors would prefer to spend all of the seigniorage on a citizens' dividend. The idea has had some tradition since the 18th century, when the governors of the British colonies in America (the later U.S.) issued sovereign paper money, called colonial bills, by way of a per-capita dividend to each taxpayer.[16]  

Basically, the currency register and, as far as it is connected to it, the central-bank balance will issue the money but not decide on its uses. In actual fact, however, the issue of money by seigniorage or by way of credit to banks is an implicit preliminary decision on the scope of the likely first uses of the money. To this extent, the creation and issue of new money and its first uses cannot entirely be kept apart.  

The deposits of the currency register in the reserve account of the central bank's balance sheet can be seen as credit at conditional notice. It can be left open here what the criteria for such notice would be – for example, whether on re-exchange of foreign currency, or on accumulated repayments of credit by the banks exceeding a certain limit to money reserves. The deposits are non-interest-bearing, in contrast to the short-term loans of the central bank to banks, because, as the accounts involved belong to one and the same central bank, payment of interest is without function here.

The register's currency account and issue account show on balance the same amount of money. Most of it will be currency in domestic use, part of it perhaps also used abroad. In terms of present monetary aggregates, the money in domestic use plus foreign-used cash correspond to M1. The reserves in M0 need not be considered here, because in terms of set-theory arithmetic – and under conditions of business-as-usual rather than ultra-loose flooding with reserves – the reserves in M0 represent a re-actively provided fractional sub-set of M1. Today's M2, M3, M4 and other such aggregates would no longer exist. Deposits in M2/M3, for example, would no longer represent temporarily deactivated bankmoney. Instead, they would explicitely be customer loans to banks. The centuries-old question of whether bankmoney is about regular or irregular deposits would thus have lost its relevance.     

In a pure sovereign money system, there is no split circulation of reserves in the interbank circuit and bankmoney in the public circuit among nonbanks. The distinction between monetary institutions (banks) and non-monetary institutions (nonbanks), too, would have lost its relevance. There would just be account money in money accounts, circulating among all the actor groups in the same way. This automatically also applies to account-related 'e-cash' cards, prepaid cards, credit cards, payment apps and similar methods of transferring money-on-account. Such devices help transfer account money but are not themselves carriers of the money. Whether there will be true mobile money stores in the future remains to be seen.

As far as sovereign account money and bankmoney coexist, previous distinctions remain relevant, more precisely the distinction between (a) bankmoney in a bank giro account, (b) fractional banking reserves in a central-bank account and (c) newly existing sovereign account money in public circulation apart from bankmoney. Sovereign account money can easily be transferred between customers, banks and other payment service providers and the central bank, including transfers between customers' bank accounts and sovereign money accounts. In the latter case, the banks' holdings of reserves are affected.

Where and how the sovereign money in nonbank use will be managed is not entirely settled yet. A 'central bank account for everyone' is unlikely. In the course of becoming national currency authorities, the central banks have abandoned conducting business and running accounts for nonbanks, except for a number of government transaction accounts. The example of Ecuador, however, shows that a 'central bank account for everyone', thus sovereign account money in public use, is feasible. In Ecuador it is called 'dinero electrónico'. Access to it, however, is possible only via special 'transaction centres'.[17]   

Without too much effort it is also possible to implement central bank accounts for general use in the form of customer transaction omnibus accounts, managed in trust by banks and other payment service providers with access to the central bank payment system. Customer transaction omnibus accounts might be affiliated with the central bank account of the payment service providers, but off their balance sheets and thus separate from the service providers' own money. The customer transaction omnibus accounts would in fact represent a part of a bank's customer current account taken off its balance sheet. It might even be possible to keep the customer-specific account numbers.

An alternative option, presumably involving more effort, is to build up a new infrastructure for individual sovereign money accounts beyond the present system of cashless interbank payments. This approach is currently pursued with the e-krona concept of the Swedish Riksbank.[18] As cash is falling into disuse in Sweden, the e-krona concept combines an account-based e-krona (sovereign account money) with a value-based e-krona (sovereign cryptocurrency), the latter acting as a continuation of cash in a modern form.

A similar combination was involved in pilot projects by both the central banks of Canada and Singapore, restricted, however, to banks and other financial institutions. A part of central-bank reserves was provided to them in the form of cryptocoins. The cryptocoins circulated between the institutions, instead of the circulation of reserves in the current real-time gross-settlement payment systems of central banks. At the end of the day, open claims were settled in reserves, while the cryptocoins were deleted. So, what was tested here is the operability of the distributed ledger and blockchain technology in payment processes rather than a change in the monetary system.[19]  

The Central Bank of Uruguay has announced running a pilot project with 10'000 users and 20 million 'peso tickets' in digital wallets accessible by a special app.[20]   

Back to top >

Changes on a central bank's balance sheet

The balance sheet of a central bank would reflect its banking business, much in the same way as the practice until now. As one substantial change, however, a central bank would no longer account for its notes in circulation and non-cash interbank reserves as liabilities to the banking sector. Instead, sovereign money would enter the central-bank balance sheet as a deposit from the currency register and would accordingly show up as liquid money on the asset side and a liability to the currency register on the liability side. The central bank's currency reserves would consist of cash, reserves/sovereign account money and perhaps sovereign cryptocurrency. Put differently, the position liabilities to the banking sector of the present split-circuit reserve system would be replaced in a sovereign money system by a position liabilities to the currency register.          

When a central bank extends short-term credit to a bank, this results in an asset swap in that an amount of liquid reserves is swapped for a credit claim, be the credit disbursed in cash, on account or crypto. The same happens when the central bank purchases foreign exchange, securities and other assets. Means in the currency reserves account are swapped, for example, for claims on foreign central banks, on financial institutions, on industrial corporations, or on domestic and foreign treasuries; the latter maybe from ways and means advances, as was usual before the introduction of the  euro, but is, incomprehensibly, interdicted now (qui bono?).    

Back to top >

 Accounting for sovereign money by banks

In a pure sovereign money system, bank accountancy would be changed in a way similar to the change in central bank accountancy. There would be no more bankmoney and thus no more overnight liabilities to nonbank customers. Equally, there would no longer be reserves in the current sense, either excess or minimum reserves, and less so a 100%-reserve on deposits that would equally be irrelevant. Sovereign money is no promise to pay but the 'real thing' itself, legal tender for the final settlement of all kinds of liabilities. Sovereign money needs no backing by reserves. Certainly, however, banks in a sovereign money system continue to need an adequate buffer of liquidity and equity, enough to avoid melting away immediately in any next crisis.             

As explained in the opening section of this paper, banks in a sovereign money system are no longer monetary institutions able to create bankmoney. They can lend, invest or spend as much sovereign money as they are able to finance by taking it in or up. The resulting bank liabilities include those
-    to customers (savings or time deposits as explicit customer loans to a bank) 
-    from self-emitted securities (bonds, bills, certificates  and other debentures) 
-    to other financial institutions from short- and long-term borrowing
-    to the central bank, in the last instance, from short-term borrowing.

In a constellation of sovereign money and bankmoney existing side by side, banks continue to act as monetary institutions with the privilege to create bankmoney on a fractional base of reserves. However, they are obliged to transfer bankmoney into sovereign money accounts of customers, which means they have to transfer reserves as they do with other banks and central-bank government accounts. Because of customer demand, most banks would find themselves constrained to offer sovereign money accounts. Should competition in the banking sector not work in that direction, perhaps because of oligopolistic corporatism in a number of countries, a little regulatory constraint would have to help.

When bankmoney is transferred from a sovereign money account to a bankmoney account (giro account), the banks obtain reserves for free. The free availability of such means for the banks will however be limited, because they need these means to make outgoing payments from giro accounts to sovereign money accounts. Further questions related to a constellation of bankmoney and sovereign money side by side cannot be discussed here. 

Back to top >

Core elements recapped

Introducing a central bank-run currency register for the purpose of creating, issuing and recording the stock of sovereign money, separate from a central bank's banking business, opens up the possibility to overcome the outdated and misleading practice of accounting for money created by central banks and banks as a liability rather than a liquid asset.

Accounting on a central-bank balance sheet for legal tender notes and reserves as a liability to the banks, reminiscent of the times when paper money was promissory notes convertible into silver coin and bullion, missed the point a long time ago. In private commercial banking, the practice of creating bankmoney as a liability to customers has resulted in an extreme situation of fractional reserve banking with particular problems, dysfunctions and injustices, the more so as the promise to pay cannot be kept by the banks, since it is literally impossible to convert bankmoney into cash or transfer the bankmoney to other banks when too many customers try to do so at once. Today, converting bankmoney into sovereign account money is impossible anyway. Reserves – that is, central-bank account money – never leaves the interbank circuit and thus never becomes the possession of a customer, but always remains in the possession of a bank, in spite of the credit law in most countries demanding that the amount granted by credit extension is fully disbursed from the creditor's assets and entered into the debtor's (not another bank's) possession. That is the legal trick, so to speak, of fractional reserve banking, which is at the root of the inherent instability and crisis proneness of finance today, including the overt non-safety of bankmoney in a crisis.

The current major response to the problems is increased equity and liquidity requirements according to the Basel rules III and maybe IV. Such measures are useful to a limited degree but will not help much as they miss the mark. They focus on the banks' balance sheets, hoping in vain for not-so-bad crises, while ignoring the fundamental role of money and the monetary system. Similarly, also a 100% reserve requirement remains caught in the twisted logic of split-circuit reserve banking while burdening banks and customers with additional costs.[22]

A systemically consistent response, by contrast, was given by the Currency School and its major principle, which is to separate money creation from bank credit, including the segregation of the customers' money from a bank's own money. This is exactly what a separate currency register does, and what the introduction of sovereign account money for everybody's use does, in future perhaps also including sovereign cryptocurrency.

The creation and issue of sovereign money through a separate central-bank currency register allows for consistent double-entry bookkeeping on the central bank's balance sheet, thereby also shaping future changes in the related accountancy practice of commercial banks, without the false and confusing identity of money and credit and without overstretching the notions of liability and equity. In particular this also relates to the transfer of genuine seigniorage to the public purse which needs to be registered neither as government debt nor as central banking equity. 

The core element here is to account for sovereign central-bank money as a liquid asset only, circulating by direct full asset transfer from payer to payee among and between banks and nonbanks alike. In connection with the lending or investing of money, there is a swap of assets, that is, a claim on money against the money.

Generally speaking, for a holder of a means of payment, that means is in fact always a liquid asset, even if it was issued as a liability of a bank (bankmoney) or of a central  bank (reserves). The present practice of accounting for bankmoney and reserves is inappropriate to the asset nature of money, which creates a truly Babylonian confusion about money and credit in the present bankmoney regime. Money thus no longer ought to be created as 'liability money'.[23] In the first place this applies to central-bank money in every form.

Bankmoney too is in fact always an asset for the holder. Should, in conclusion, banks be required to account for incoming bankmoney by making a full deduction from the incoming reserves until a respective customer makes an outgoing payment, this would come close to the beginning of a sovereign money system. Still missing would be the transfer of the reserves into a separate sovereign money account or into a customer transaction omnibus account or whatever arrangement there may be to ensure that bank customers are set in full possession of their money rather than reserving the money for the banks and letting bankmoney circulate as a rather empty promise to pay.[24] 

Back to top >


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

Benes,  Jaromir / Kumhof, Michael. 2012. The Chicago Plan Revisited, IMF Working Paper, 12/202, August 2012. Revised draft February 2013.

Gudehus, Timm. 2013. Geldordnung, Geldschöpfung und Staatsfinanzierung, Zeitschrift für Wirtschaftspolitik, Vol. 62, No. 2, 2013, 194–222.

Gudehus, Timm. 2016. Neue Geldordnung, Wiesbaden: Springer Gabler.

Hixson, William F. 1993. Triumph of the Bankers, London/Westport CT: Praeger.

Huber, Joseph. 2017. Sovereign Money – Beyond Reserve Banking, London: Palgrave Macmillan.

Huerta de Soto, Jesús. 2006. Money, Bank Credit and Economic Cycles, Auburn AL: Ludwig von Mises Institute.

Jackson, Andrew / Dyson, Ben. 2012. Modernising Money, London: Positive Money.

Jackson, Andrew / Dyson, Ben. 2013. Sovereign Money – paving the way for a sustainable recovery, London: Positive Money.

Köhler, Michael. 2015. Humes Dilemma – oder: Das Geld und die Verfassung. Geldschöpfung der Banken als Vermögensrechtsverletzung, Berlin: Duncker & Humblot.

Mayer, Thomas [Swiss Vollgeld-Initiative]. 2017. Bilanzierung von Bankengeld und von Vollgeld, available at

Mayer, Thomas [FvS Research Institute, former chief economist of Deutsche Bank]. 2014. Die neue Ordnung des Geldes, München: FinanzBuch Verlag.

O'Brien, Denis Patrick. 2007. The Development of Monetary Economics, Cheltenham: Edward Elgar.

Schemmann, Michael. 2012. Accounting Perversion in Bank Financial Statements, IICPA Publications.   

Schemmann, Michael. 2015. Putting a Stop to Fictitious Bank Accounting, IICPA Publications.

Schularick, Moritz / Taylor, Alan M. 2009. Credit Booms Gone Bust. 1870–2008, National Bureau of Economic Research, Working Paper 15512. Reprinted in: American Economic Review, Vol. 102, No. 2, April 2012, 1029–61.

Sigurjonsson, Frosti. 2015. Monetary Reform – A Better Monetary System for Iceland, commissioned by the Prime Minister of Iceland, Reykjavik.

Sveriges Riksbank. 2017. The Riksbank's E-Krona Project, Report 1, Stockholm, September 2017.

Tobin, James. 1987. The Case for Preserving Regulatory Distinctions, Proceedings of the Jackson Hole Economic Policy Symposium, vol.30, Feb 1987, no.5, 167–205.

Zarlenga, Stephen. 2002. The Lost Science of Money, Valatie NY: American Monetary Institute.

Back to top > 


[1] Tobin 1987. 

[2] Except for Bech/Garratt 2017.

[3] For the criteria of issuing sovereign money see, or
Also cf. Huber 2017, 157 et seq.   

[4] Schularick/Taylor 2009, 2, 5ff., 29.

[5] The volumes of MMF-shares are important, amounting to a third of M1 (EU, UK) to two and a half times M1 (USA).      

[6] IFRS= International Financial Reporting Standards by the IASB = International Accounting Standards Board (EU und UK). GAAP = Generally Accepted Accounting Principles by the FASB = Financial Accounting Standards Board (USA).

[7] The idea of just formal credit, with no maturity and free of interest, was also set forth by Keynes, named perpetual zero-coupon consols, apparently inspired by the British bond issuances of the 18th and 19th centuries. These bonds were time and again re-consolidated into new bond packages issued at lower interest each time. Jackson/Dyson 2013, 18 et seq., have taken up Keynes's consols idea. As for myself, I had supported that approach until the first edition of the book Monetäre Modernisierung, 2010, replaced in subsequent editions with the idea, as explained below, of treating sovereign money on a central-bank balance sheet in analogy to the issuance of coin – in turn extended in the present paper by the approach to a central-bank currency register. Sigurjonsson 2015, 81, thinks that both the consols approach (liability approach) and the coin or equity approach are applicable.   

[8] In contrast to a commonly held belief, minimum reserves do not normally act as a liquidity safety buffer, less so as a policy instrument for controlling the overall volume of bank credit according to unreal models of a credit multiplier. Only marginally and temporarily can minimum reserves be put in as excess reserves if a respective bank on average complies with the minimum reserve requirement over the entire reference period.

[9] In Germany, for example, this is stipulated in § 488 BGB (the Civil Law Code) and its general interpretation.

[10] Schemmann 2012, 2015. Most neo-Austrian economists also consider fractional reserve banking to be fraudulent, for example Huerta de Soto 2006, chs. 2 and 3, or Köhler 2015.                

[11]  Cf. Jackson/Dyson 2012, 210, 311–21. Benes/Kumhof 2012, 6. Gudehus 2013, 199–207; 2016, 9–11. Mayer [FvS] 2014, 146–61. Mayer [Swiss sovereign money initiative], late 2017. Sigurjonsson 2015, 81. Huber 2017, 167–9.

[12] O'Brien 2007, 93–154.

[13] The state prerogative of coinage most often still rests with the national treasuries, as a remnant of feudal times long gone. The prerogative should have been conferred to the national banks long ago, which implies that these are nationalised monetary authorities, preferably no longer run as joint-stock companies and certainly not with private institutions and persons as stock co-owners, least of all banks and other financial corporations.  

[14] For more on the subject of monetary financing, see 

[15] In a transition from bankmoney to sovereign money, two categories of seigniorage can be distinguished: the regular seigniorage from current money creation and the one-off transition seigniorage or conversion seigniorage from substituting sovereign money for bankmoney. The amount of one-off conversion seigniorage would be very high in comparison, in fact amounting to the stock of previously existing bankmoney. The question of what to do with that much money needs to be dealt with separately. Among the obvious options is paying down the major part of public debt or putting the money into any kind of public investment.

[16]  Hixson 1993, 53–58. Zarlenga 2002, 361–86. The approach lived on, for example as national dividend or citizens' dividend in the writings of C.H. Douglas and as 'free money' tax allowance in S. Gesell. The idea continues to be supported by present-day monetary reformers. Cf. Mayer [FvS] 2014, 148ff. Huber 2017, 160–63.  

[17] Bech/Garratt 2017, 62.

[18] Sveriges Riksbank. 2017.

[19] Bech/Garratt 2017, 61, 66.

[20] Central Banking, 6 Nov 2017, Uruguay launches digital currency. News, Nov 6, 2017, Uruguay to launch digital currency. 'Not bitcoin' it stresses.

[21] For more cf.  

[22] See

[23] According to the term 'Passivgeld' by Th. Mayer [FvS] 2014, 22ff.  

[24] A fiduciary account at the same or another bank does not fulfill the same function, as fiduciary balances remain an item on a bank's balance sheet, subject to fractional reserve banking.  

Back to top >