How to modernise central bank balance sheets: no notes

Unlock the Editor’s Digest for free

Roula Khalaf, Editor of the FT, selects her favourite stories in this weekly newsletter.

David Bholat is Director of Financial Services at Faculty and a former senior manager at the Bank of England

John Maynard Keynes once quipped that gold was a “barbarous relic” unfit to function as money in a modern economy. The same might be said today of the way central banks account for banknotes on their balance sheets, precisely because they are no longer redeemable for gold.

Almost everyone still recognises banknotes as assets, even without metallic backing. Almost everyone, but not the central bank that produces them. The Bank of England’s balance sheet is illustrative.

The classification of cash as a liability hearkens back to a time when banknotes promised to pay bearers gold on demand. But this classification has lost currency, so to speak, now that notes are redeemable only for the same or a similar claim on the central bank.

This accounting quirk has recently attracted academic scrutiny. A paper written by Michael Kumhof of the BoE with several co-authors makes the case that classifying banknotes as liabilities is anachronistic. Instead, they argue for treating banknotes as a component of capital.

A benefit from doing so is that it would plug the capital shortfall some central banks are facing from losses or writedowns in the value of their assets post-pandemic. In the UK, current rules obligate the Government to recapitalise the BoE if its capital falls below a capital floor. Recognising banknotes as part of the BoE’s capital would create an additional buffer before taxpayer liability kicks in.

The trouble is that banknotes are not fit for absorbing losses. Withdrawing or writing-down notes in circulation could generate deflation, with potential adverse effects on employment and output. At the limit, it could cause a crisis of confidence in a central bank.

There are also conceptual challenges with classifying banknotes as equity from a purely accounting point of view. Banknotes do not conform to the definition of equity under IFRS, as they don’t entitle their bearers to receive the residual interest in the net assets of the central bank if it were liquidated. Unlike ordinary shares, banknotes confer no rights to vote on matters related to central bank management. And unlike ordinary or preference shares, bearers of banknotes aren’t entitled to dividends.

Alternatively, banknotes could be classified as a hybrid instrument countable as capital, similar to subordinated debt. But the underlying economics of a ‘claim’ which generates no return, ever, hardly makes it worth the name.

The blinkers of balance-sheet accounting shouldn’t obscure another option, which is to remove banknotes entirely from the statement of financial position.

The Royal Mint’s approach to coins offers a case in point. The Mint does not recognise the currency it manufactures on balance sheet. Instead, revenue and costs from currency are recognised through the Mint’s income statement.

The Mint’s approach sidesteps the awkward classification of currency as either debt or equity. Central banks could take a similar approach. The difference between the nominal value of banknotes and their cost of production would flow through as profit (seigniorage) on income statements. While banknotes would no longer appear on balance sheets, they would still bolster them, as the seigniorage could be retained as earnings to boost capital.

The key conceptual shift is to rethink banknotes, not as debt or equity, but as the central bank’s product. The demand these products fulfil is facilitating economic exchange, transfers between states and citizens, and social transactions between individuals. Like many products, banknotes come with implicit warranties (the replacement of worn notes with new ones), services (the option to swap physical notes for digital reserves), guarantees (the central bank’s promise to retain banknotes’ real value through policies that promote price stability), and licensing agreements (no counterfeiting). 

Removing banknotes from central bank balance sheets would require additional accounting adjustments to restore its balance. One possibility is to discharge an equivalent amount on the assets side. For example, BoE notes in circulation are backed by securities including gilts. Discharging these backing assets would shrink public debt and the central bank’s balance sheet, providing one mechanism for pursuing quantitative tightening.

Abolishing the antiquated practice of classifying banknotes as liabilities could also set the stage for eliminating another anachronism in the way BoE accounts are presented. Currently, the BoE reports two sets of financial statements (excluding the PRA): one for the Banking Department and another for the Issue Department. The Banking/Issue Department split has little to do with how the BoE operates today.

Instead, it reflects an outdated accounting convention rooted in the 1844 Bank Charter Act. That Act segregated the BoE’s role in creating currency (the Issue Department) from its role discounting commercial paper and advancing loans (the Banking Department). The purpose of ringfencing the accounts this way was to anchor the issuance of banknotes to the BoE’s gold reserves.

Now that gold has gone, the BoE’s financial statements could be consolidated. Retaining the distinction between the Issue and Banking Department today serves no purpose, except that it still bears crucially on profit distribution. While only a portion of the Banking Department’s profits are potentially payable as dividends to the government, all profits earned by the Issue Department are paid into the Treasury’s account with the BoE. Last year, the Issue Department’s net profit was about £4.2bn. Retaining those earnings would have roughly doubled the BoE’s net worth, boosting its loss-absorbing capital more directly than reliance on recapitalisation guarantees from the government.

Of course, the fiscal and economic implications of changing banknote accounting require more deliberation and discussion than is possible here. Several knotty issues would need to be worked through, including the appropriate treatment of other forms of central bank money such as reserve accounts and CBDCs. Whether it’s even worth the fuss is also debatable. The use of cash as a share of transactions fell to a record low last year. To paraphrase Keynes once more, the euthanasia of banknotes may be more likely than any revolution in the accounting for them.