By Nobukazu Ono and Álvaro Pina, OECD Economics Department
After recording sizeable gains from quantitative easing (QE) over most of the past decade, several central banks have recently reported financial losses. Further losses are likely in the coming years. As discussed in the latest OECD Economic Outlook (OECD, 2023), these losses have long been anticipated, as past gains and current losses are two sides of the same coin. This is also the case for the adverse impact of losses on the public finances, which comes after several years of significant central bank remittances to Treasuries. Nonetheless, losses and the potential need for recapitalisation from the government can still undermine the credibility and perceived independence of central banks. This calls for transparency and clear communication in the presentation of losses.
Why are central banks making losses?
Losses largely reflect the large balance sheets built up during an extended period of QE and the effects of the recent rises in policy interest rates. QE created a sizeable mismatch in the maturity of central bank assets and liabilities. On the liability side, central bank reserves (mainly commercial bank deposits) rose sharply. The remuneration on these is closely linked to policy interest rates, and has thus risen rapidly, to the benefit of commercial banks. In contrast, on the asset side, QE purchases were mostly long-term fixed-rate bonds that generate a relatively stable stream of income. When policy interest rates were at or close to the zero lower bound, the balance of these two sets of payments generated gains for central banks. Even as policy rates were raised through 2022, their impact on whole-year net interest income was still relatively mild (Figure 1), especially where most rate increases took place towards the end of the year, as in the euro area. However, larger impacts are likely in 2023 and 2024 (Anderson et al., 2022; De Nederlandsche Bank, 2022).
Higher interest rates also reduce the market value of securities. Valuation losses may thus arise, though this depends on the accounting frameworks and asset sales decisions of central banks. For instance, the Federal Reserve and the Bank of Japan account for securities held for monetary policy purposes using amortised cost. This means that valuation changes do not affect profits unless securities are sold, which has not been the case so far. Eurosystem accounting guidelines, also followed by Sweden, allow central banks to value securities held for monetary policy purposes at either amortised cost or the current market price. While euro area national central banks have generally opted to use amortised cost, the Riksbank has adopted market pricing, making a significant loss apparent in 2022. Mark-to-market accounting brings forward loss recognition, as also illustrated by Australia, Canada, New Zealand, Switzerland and the United Kingdom. In Switzerland, the central bank loss was unusually large in 2022, at 17 per cent of GDP, but these losses stemmed largely from changes in the domestic currency value of foreign exchange reserves, including foreign securities.
Do losses hamper fiscal or monetary policy?
Central bank losses affect the public finances by reducing or ending central bank payments to the Treasury in the form of income taxes or remittances. Moreover, reverse cash flows (i.e., payments from the Treasury) may occur if central banks are entitled to be compensated by the government for certain losses, such as QE-related losses. For instance, in the United Kingdom, the Bank of England Asset Purchase Facility (APF), through which QE asset purchases were conducted, is fully indemnified by the Treasury.
The annual impact on the general government fiscal balance should in general be modest: over 2015-21, revenues from the central banks considered in Figure 1 rarely exceeded, and were often far below, 0.5% of GDP per year. However, even if small, the negative effect on fiscal balances could be protracted, as losses may persist for several years. Even after the central bank returns to profit, some time may elapse before remittances to the Treasury resume.
Central bank losses are not an indication of a policy error and need not hamper the effectiveness of monetary and financial policies. The policy mandates of central banks include price stability and financial stability, but not profit maximisation. Their current losses, as well as their earlier gains from QE, are a by-product of policy actions designed to help achieve their mandates. Moreover, central banks are not subject to capital adequacy requirements or bankruptcy procedures and can operate effectively even with negative equity, as the central banks of Chile, the Czech Republic, Israel and Mexico have done over several years (Bell et al., 2023).
However, losses or negative equity can pose communication challenges. For instance, some policy decisions, such as retaining rather than selling government bonds, could be misinterpreted as being motivated by a desire to contain losses rather than as actions to pursue specific policy mandates. This would reduce central bank credibility. Likewise, financial flows from government, including actions to strengthen central bank capital positions, could be misperceived as being inconsistent with central bank independence. This underscores the importance of clear communication about the reasons for losses and of a transparent framework for financial flows between the central bank and the government.
Figure 1: Several central banks now report negative profits
Note: Accounting approaches and financial years differ across countries and data should therefore be compared with caution. ‘Net gain from asset valuation’ includes realised gains/losses and, in some countries, unrealised ones. For the euro area, the chart shows the consolidated result of national central banks. The euro area figure for 2022 is based on the 17 member banks who had published results by 29 June 2023. For the United Kingdom, the chart shows the consolidated result of the Bank of England and the Asset Purchase Facility. See Box 1.3 in OECD (2023) for further details.
Source: Board of Governors of the Federal Reserve System; Bank of Japan; national central banks in the euro area; Bank of England; Bank of Canada; Reserve Bank of Australia; Swiss National Bank; Sveriges Riksbank; Reserve Bank of New Zealand; OECD Quarterly National Accounts database; and OECD calculations.
References
Anderson, A., P. Marks, D. Na, B. Schlusche, and Z. Senyuz (2022), “An Analysis of the Interest Rate Risk of the Federal Reserve’s Balance Sheet, Part 2: Projections under Alternative Interest Rate Paths,” FEDS Notes, Washington; Board of Governors of the Federal Reserve System, 15 July, https://doi.org/10.17016/2380-7172.3174.
Bell, S., M. Chui, T. Gomes, P. Moser-Boehm, and A. P. Tejada (2023), “Why are Central Banks Reporting Losses? Does it Matter?,” BIS Bulletin No 68, February, https://www.bis.org/publ/bisbull68.pdf.
De Nederlandsche Bank (2022), “Letter regarding DNB’s capital position – September 2022,” 9 September, https://www.dnb.nl/media/p4igu5fc/letter-regarding-dnb-s-capital-position-september-2022.pdf.
OECD (2023), OECD Economic Outlook, Volume 2023 Issue 1, OECD Publishing, Paris, https://doi.org/10.1787/ce188438-en.