Why I’m not concerned about the billions of central bank “losses”

Maria Demertzis*

How well are central banks doing? Right now, they are making losses but, unlike most large institutions, we should not assess their performance on annual profit and loss. In fact, central banks’ annual losses should not even be considered relevant in assessing how well they perform their tasks.

These losses are the consequence of central banks having large balance sheets while interest rates are simultaneously high. According to a recent IMF paper, central banks in the euro area are expected to make €55bn in losses between 2023 and 2024, equivalent to 0.5% of the euro area’s GDP.

The Bundesbank and the Banque de France are estimated to make €26 bn and €17 bn losses in 2023, respectively. The US Federal Reserve is similarly expected to record losses of $100bn in 2023 – which could possibly peak at around $200bn by 2025.

Are these losses a problem?

The short answer is no. It would be a mistake to look at a central bank balance sheet, specifically its income statements, as an ‘outcome’, the way we do for any ‘firm’. Balance sheets are tools for central banks to provide liquidity to the financial system and, since the global financial crisis, to also carry out monetary policy.

The operational costs of a central bank, such as salaries and expenses, are a very small component of their income and expense account and therefore of no relevance to a country’s financial system.

Central banks are neither required to make profits nor do they somehow benefit from them when they do, as profits are returned to their shareholders – namely the state.

But equally, if central banks make losses, they can keep them on their books as negative equity. This will have no implications for their ability to do their job: provide liquidity, change the interest rate, or apply quantitative easing if they need to. These behemoths do not need profits to operate, nor do they run out of liquidity.

Taking a closer look at QE

When central banks started quantitative easing, they bought large amounts of government bonds of their respective sovereign that are housed on the asset side of their balance sheets.

Two things changed as interest rates increased over the past couple of years. The first is that central banks had to pay larger amounts of interest on the part of their liabilities, namely the reserves that commercial banks hold at central banks, which dwarfed the coupon they received on all those bonds they bought.

Second, an increase in interest rates meant a fall in the price of their bond holdings on the asset side, which in principle leads to valuation losses.

As central banks in the euro area do not value these assets at market price, these losses are not realised for as long as they hold these assets are held to maturity.

They are therefore irrelevant. But for some central banks that mark those assets to market, like the Swiss National Bank and the Swedish Riksbank, valuation losses are realised and can contribute to significant losses.

Earlier this month it was reported that the Riksbank, the world’s oldest central bank, had asked for a taxpayer “recapitalisation” worth $7 billion to plug losses racked up in the past two years.

If such losses were sustained for a long period of time, then questions would arise regarding central banks’ ability to finance their own activities. Accumulating negative equity could risk central bank independence and erode public trust in their ability to perform their tasks without a government capital injection. Governments are within their right to ask the question of whether they will be called to provide such injections.

But beyond that, there is also an issue of incentives. Ultimately, if increasing interest rates is part of the reason why losses are sustained, there is an issue as to whether central banks can be trusted to use the instrument as needed to achieve price stability.

Wouldn’t the prolonged coexistence of large balance sheets and high interest rates distort the central bank’s willingness to tighten further, if there was a need? Surely a central bank’s reputation is damaged if they ask for a taxpayer contribution to their capital? It depends. In the case of the Riksbank, it was obliged to ask for a recapitalisation once losses exceeded a set threshold under a new law that came into force this year. It was not a decision led by the Riksbank’s risk management experts, rather the long arm of the law.

The same IMF paper I cited before showed that the current and expected losses for the next couple of years for a number of central banks are temporary, and can be fully recuperated by 2027. As argued in this 2018 Bruegel paper, so long as central banks have a balanced position across time, there is no issue.

Also, central banks make appropriate provisions to manage portfolio risks from one year to the next. All of this means that the losses observed will not jeopardise any central bank’s ability to finance their own activities.

Could these losses have been avoided?

Central banks could have avoided these losses only if they hadn’t applied quantitative easing to stimulate the economy first, and then increased interest rates to tame inflation. But this also means they would not have done their job of safeguarding price stability.

The slow speed at which assets mature and balance sheets are unwound implies that there will be periods in which large balance sheets will have to co-exist with high interest rates. The maturity profile of their assets is known, and central banks can therefore identify the timing for which they need to provision against future losses. This is an integral part of their risk management.


*Maria Demertzis is a Senior fellow at Bruegel and part-time Professor of Economic Policy at the School of Transnational Governance at the European University Institute in Florence. The article was originally published on the FT’s Banking, Risk & Regulation,  also posted by Bruegel and on the blog of the Cyprus Economic Society.

Related Posts