cnBlog – A look at selected central banks’ balance sheets and finances
Over the past 15 years, the size of central banks’ balance sheets has increased considerably, reaching record levels in many cases around 2022. Since then, balance sheets have been shrinking in most cases, but are still high from a historical perspective. The combination of large balance sheets and a sharp monetary policy tightening led to a significant deterioration in their profits and, in some cases, also to a shift in their equity to negative values. For central banks, however, their objective is not to generate profits, but to fulfil their price and financial stability mandate.
Published in Central bank monitoring – December 2024 (pdf, 695 kB)
Since the global financial crisis, many central banks have taken unconventional measures (e.g. quantitative easing, various liquidity-providing programmes, exchange rate commitments, yield curve control). The common denominator in these measures was a sizeable increase in these central banks’ balance sheets. Further substantial growth in central banks’ balance sheets was due to measures adopted after the outbreak of the Covid-19 pandemic. Central banks’ balance sheets have been shrinking in most cases over the last approximately 2–3 years, but are still high by historical standards. The combination of the large volumes on balance sheets and sharp monetary policy tightening in recent years has in many cases led to a significant deterioration in central banks’ finances. After a brief theoretical introduction, this article focuses on an overview of trends in selected central banks’ balance sheets, describes the impact on their finances, examines practices in profit distribution and discusses some specifics of individual countries.[1]
Effect of central banks’ balance sheets on their finances
Monetary policy settings and the size of a central bank’s balance sheet are linked in several ways to its finances. In the case of central banks that purchased large amounts of government bonds in the past (often with a low yield and long-term maturities – and hence with high sensitivity of bond prices to changes in interest rates), the sharp increase in interest rates in recent years led to a decline in bond prices and therefore a decline in the value of these central banks’ assets. Central banks’ asset purchases were financed on the liabilities side by an increase in the volume of reserves for which central banks pay interest to commercial banks as a part of monetary policy in an environment of excess liquidity. Given the high volume of reserves, a rise in interest rates implies a significant increase in monetary policy costs. The radical tightening of monetary policy around 2022 thus worsened the finances of many central banks through both these channels. Accounting standards, which differ across countries, also have an effect on the reported profit or loss – some central banks value their asset holdings according to their mark-to-market value, while others use their book value (in such case, the aforementioned fall in bond prices caused by a rise in interest rates does not lead to an accounting loss until the loss is realised by selling the bonds – although as a part of quantitative tightening some central banks do not keep the bonds to maturity, but are gradually selling them and so incurring losses).[2]
In the case of central banks where a large proportion of assets are accounted for by foreign exchange reserves, a high balance sheet also implies greater sensitivity to exchange rate movements of the domestic currency – in the event of a strengthening of the domestic currency, the value of an asset denominated in foreign currency when converted into the domestic currency also decreases. Under otherwise identical conditions, a strengthening of the domestic currency thus leads to a worse result for the central bank, whereas a weakening of the domestic currency conversely improves it.
The effect of specific monetary policy actions on a central bank’s profitability cannot be interpreted as an indicator of whether these actions were correct or not. For example, the aforementioned sharp increase in rates in the recent past had a visibly negative impact on central bank profitability and, if central banks had left rates close to zero instead, they would have achieved better results – but it would have meant forgetting the fight against high inflation.[3] The central banks’ aim is to achieve price and financial stability (including, for example, low unemployment in some countries) and not profit. Moreover, central banks cannot become insolvent and can also function with negative equity in the long term (as evidenced by a number of examples in practice).
On the other hand, a central bank’s profit and loss cannot be regarded as completely irrelevant. Some central banks[4] have regularly contributed from their profits to their country’s central government budget in the past. If they were to make a loss, the central government budget would lose these funds. Other countries must cover some of the losses under the legislation in force, or a central bank must be recapitalised if its capital falls below a set threshold. Even in countries where the payment of central bank profits to the central government budget was not common practice in the past and the central bank does not have to be recapitalised under the law – and therefore the central bank’s losses do not have an immediate impact on the central government budget – there is a long-term impact, as even in these countries the central bank could potentially contribute to the central government budget in the future if a sufficiently high level of equity was achieved and a current loss delays this moment in the future.
A debate on the extent to which the low or negative value of equity poses a risk to fulfilling the price stability mandate is taking place in specialised literature. Bell et al. (2023) argue that a central bank’s losses do not jeopardise the central bank’s ability to meet its objectives. By contrast, they may sometimes be a necessary price of achieving them. However, it would be risky if the poor financial position of a central bank were to cause political pressure on the central bank or reduce its credibility. Some extreme cases, where (depending on the institutional set-up of financial transfers between the central bank and the government) the sustainability of a central bank’s finances may not be consistent with achievement of the inflation target, are analysed, for example, by Hall and Reis (2015). According to the authors, such a situation is possible in theory but unlikely in practice.
Balance sheets of key central banks in the Euro-Atlantic area[5]
Before looking closely at the selected central banks’ balance sheets in recent years, we will look briefly at very long-term trends, specifically using the example of the Bank of England. This is because research data for this institution are available from 1697 (although statistics that are several hundred years old should naturally be treated with considerable caution). A look at Chart 1 shows that the BoE’s balance sheet has seen booms and declines (at least in relation to GDP) in the past. However, it can also be seen that the speed and size of the growth in the BoE’s balance sheet since the global financial crisis is unprecedented.[6]
Chart 1 – Bank of England balance sheet 1697–2019
Note: These are research data, not official statistics.
Source: Bank of England research datasets
To make a better comparison, in the text below we present the individual central banks’ balance sheets in relation to the relevant economy’s nominal GDP. By definition, this ratio is affected by (nominal) economic growth in addition to the absolute size of the balance sheet. In periods when the balance sheet fell as a share of GDP in the past, this decline was usually due to growth in the denominator (GDP), while the balance sheet volume remained more or less constant.[7] However, many banks have also reduced their balance sheets in absolute terms, especially in the last 2–3 years. An illustration can be provided by the example of the Fed in Chart 2, which compares the nominal value of its balance sheet and its balance sheet-to-GDP ratio. This phenomenon was most apparent in 2014–2017, when the Fed (after the previous several waves of QE in response to the financial crisis and its impacts) did not make net asset purchases, but merely reinvested maturing assets. The nominal value of the balance thus remained virtually constant, while the ratio of the balance sheet to GDP gradually decreased. Between the end of 2017 and the start of the Covid-19 pandemic, and especially from 2022 to the present, the size of the balance sheet in absolute terms fell as a result of quantitative tightening.
Chart 2 – Fed’s nominal balance sheet and compared to US GDP
Source: BIS
In the past, the Fed regularly sent its profits to the US Treasury, amounting to tenths of a per cent of GDP a year (between 2011 and 2021 the total was more than USD 920 billion, for details see this blog article on the St. Louis Fed’s websites). A change occurred in September 2022, when the Fed posted a loss and stopped sending funds to the Treasury. Instead, the Fed created a deferred asset on its balance sheet to cover losses. After returning to profitability (which is expected in 2025 according to this year’s projections), the Fed will first cover this deferred asset and subsequently be able to resume payments to the Treasury.
Before the financial crisis, the Fed implemented its liquidity-providing monetary policy. After QE was introduced, however, it had a liquidity surplus as a result of an increase in liquidity. It is planning to stay there for a long time, so no return to the levels from the start of the century can be expected, despite a continuing decline in the Fed’s balance sheet. Before the financial crisis, the Fed operated under what it calls a scarce reserves regime. Following a significant increase in liquidity in previous years, it operated under an abundant reserves regime and its long-term intention is to operate under an ample reserves regime in which the amount of reserves just covers demand for them. However, there is some uncertainty about the question of how the specific volume of reserves (and the corresponding size of the Fed’s balance sheet) corresponds to the ample reserves regime. The Fed thus has no specific target size for its balance sheet but continuously monitors and assesses market conditions and tries to avoid any problems with liquidity in the market. Since June this year, the Fed has reduced the monthly decline in its balance sheet from a maximum of USD 95 bn to USD 60 bn.
For the euro area, it is appropriate to speak of the consolidated balance sheet of the Eurosystem, made up of the ECB and the national central banks of euro area Member States. Although decision-making on asset purchases and other measures is up to the ECB’s Governing Council, individual national central banks also contributed to the implementation and purchased assets for their balance sheets. A wide range of programmes had a significant impact on the gradual increase in the Eurosystem’s overall balance sheet over time – ranging from measures taken shortly after the onset of the global financial crisis to classical quantitative easing (from 2015 the APP programme and after 2020 also the pandemic-related PEPP) and targeted longer-term refinancing operations (TLTRO). Since 2022, the Eurosystem’s balance sheet has been shrinking, not only as a percentage of GDP (see Chart 3), but also due to the quantitative tightening and repayment of TLTRO loans. As in the case of the Fed, the previous increase in the Eurosystem’s balance sheet brought the euro area into a liquidity surplus regime. This led, among other things, to the ECB’s key rate becoming the deposit rate, at which most liquidity is absorbed (while the importance of the formerly main MRO rate decreased).
Chart 3 – Balance sheet size relative to GDP (%)
Source: BIS
The ECB distributes part of its profits to its reserve fund, while the rest is distributed to the euro area national central banks in accordance with their paid-up shares. The subsequent distribution of national central banks’ profits to individual governments is a matter of national concern – practice differs across euro area countries.[8] Over the past two years, however, the ECB has paid out no profit, or rather has not generated one. It was able to cover the loss made in 2022 by releasing the reserve for financial risks, so the resulting profit was zero, but in 2023 it made a loss for the first time since 2004, reaching almost EUR 1.3 bn (and if EUR 6.6 bn from the reserve for financial risks had not been used, the loss would have been higher). The ECB also expects to continue making losses in the next few years, after which it should return to profitability.
As regards the assets purchased under quantitative easing, a specific system is in place in the United Kingdom. The Bank of England implemented QE through the Asset Purchase Facility (APF) programme, which was implemented by a separate institution (subordinated to the BoE) set up for this purpose in early 2009. The BoE also concluded an agreement (deed of indemnity)[9] with the Treasury to transfer APF profits to the Treasury, and losses are also to be covered by the Treasury. Originally, payments for these profits or losses were to be settled on a one-off basis after the end of the APF programme, but as its likely duration lengthened, it was agreed in 2012 to settle at a quarterly frequency. In the past, this meant payments from the APF to the central government budget,[10] but from October 2022 onwards the situation was reversed after the increase in interest rates. The Treasury now has to cover losses in the APF programme. According to projections published this November, the Treasury will have to finance APF losses in the years ahead as well. The net present value of past and expected future cash flow is estimated at between GBP 50 and 95 billion paid out by the Treasury to cover losses from the APF.
Balance sheets of selected non-EU central banks with inflation targeting
Chart 4 – Balance sheet size relative to GDP (%)
Note: Attention is drawn to the double y-axis relative to Charts 3 and 5.
Source: BIS
The Norges Bank’s balance sheet has long been close to 20% of Norwegian GDP (see Chart 4). The NB did not undertake quantitative easing and, with some exceptions (foreign exchange interventions and other measures after the outbreak of Covid-19), also did not take further steps with a significant impact on the size of the balance sheet. Although the NB recorded a loss in 2022, it made several times higher profits the following year and has regularly contributed to the central government budget in recent years (including 2022).
Foreign exchange interventions have long been a key factor in the evolution of the SNB’s balance sheet. In 2011–2015, the SNB maintained a publicly announced minimum exchange rate of the franc against the euro, which entailed purchases of foreign currency. However, it did not operate in the foreign exchange market only in these years – it intervened to weaken the franc between 2009 and 2021, which led to a gradual increase in the balance sheet to almost one and a half times the annual Swiss GDP. At the end of 2022, by contrast, the SNB started to intervene in favour of the franc in response to a surge in inflation and sold foreign currencies, which led to a reduction in its balance sheet over the last two years, which, however, is one of the largest in an international comparison. The SNB’s balance sheet growth also meant greater volatility in its profit/loss. The SNB has a rich history of sending part of its profits to the federal government and Swiss cantons. Nevertheless, it recorded a record loss in 2022 and remained in the red in 2023, albeit to a much smaller extent,[11] so it had to suspend payments to the government and cantons.
Although New Zealand’s RBNZ and Canada’s BoC engaged in quantitative easing after 2020, their balance sheets are among the smaller ones in an international comparison. Like the BoE, both banks have an agreement to cover the losses arising from the asset purchase programme (indemnity). In New Zealand, the minister of finance decides on the payment of profits to the central government budget, based on a proposal from the RBNZ. After two years without paying out profits, the RBNZ paid out just under NZD 600 bn to the central government budget in financial year 2023/2024. The Bank of Canada has not contributed to the central government’s budget in recent years, as its equity turned negative in 2022 and stood at around -CAD 5.8 bn at the end of 2023.
For comparison with the countries regularly monitored in Central Bank Monitoring, Chart 4 also shows the balance sheet of the Japanese BoJ, as Japan was a pioneer of unconventional monetary policy. The BoJ was the first central bank to launch quantitative easing in 2001, even before the global financial crisis, and it pursued a yield curve targeting policy from 2016 to this year. As a result, the BoJ’s balance sheet-to-GDP ratio is very high – there has long been no major difference between it and the SNB’s balance sheet, but the balance sheets of the BoJ and the SNB are more than double those of the other central banks monitored. Unlike many other central banks, the BoJ has not made losses in recent years and regularly sends its profits to the government.
Selected central banks of inflation-targeting EU countries
The Swedish Riksbank implemented QE after 2015 and again after Covid-19 started. However, its balance sheet increased less significantly compared to many other central banks and has gradually shrunk since 2022 as a result of quantitative tightening (see Chart 5). However, asset purchases had a clear impact on the Riksbank’s finances and it made a loss of around SEK 80 bn in 2022 following a rise in interest rates and revaluation of bond holdings, which brought its own equity into negative territory. According to an act effective from last year, the Riksbank has to submit a recapitalisation request to parliament if its equity falls below a certain threshold. This was done in April this year and it also pointed out that its 2024 economic results would be very sensitive to future interest rate and exchange rate movements, so the value of equity might soon drop again. The Riksbank subsequently received a capital injection in September, although at a lower level than originally requested.[12]
Chart 5 – Balance sheet size relative to GDP (%)
Source: BIS
The Hungarian MNB’s balance sheet has been relatively volatile over recent decades and it experienced more marked growth during the financial crisis and then as a result of quantitative easing after the onset of the Covid-19 pandemic. The MNB made a loss in 2021–2023, leading to a gradual decline in equity, which turned negative in 2023. This would have entailed a need for the central bank’s recapitalisation in the past, but after last year’s amendment to the law the situation is assessed from a longer-term perspective. The current situation has been assessed in such a way that recapitalisation is not necessary for the time being and no central government budget funds were therefore provided to the central bank.
The size of the Polish NBP’s balance sheet is relatively stable compared to other countries, although a certain increase in 2020 due to quantitative easing can be seen. Given the significant role of foreign currency assets, the zloty’s exchange rate has a significant influence on the NBP’s finances. Its appreciation was the main factor underlying the fall in the NBP’s equity into negative territory in 2023.
The CNB’s balance sheet increased significantly during the exchange rate commitment period in 2013–2017, especially at the end of the commitment. The sale of foreign currency reserves, whereby the CNB prevented excessive fluctuations of the CZK exchange rate, had the opposite effect in 2022. The programme of sales of part of the income on foreign currency reserves was renewed in August 2023, slowing the growth in the balance sheet. As the bulk of the CNB’s assets are foreign currency reserves, the CZK exchange rate has long had a significant effect on its results. The CNB has long been functioning with negative equity – since 2000 it has only been positive in 2014–2016. The same as many other central banks, the CNB made a significant loss in 2022, whereas in 2023 it made a profit of around CZK 55 billion, which brought its equity to around -CZK 426 bn. Owing to negative equity, the CNB does not send funds to the central government’s budget. Any future profits will first be used to settle the accumulated loss; after covering it, the CNB would use the additional profit initially to replenish the reserve fund and the remaining profit would be sent to the central government’s budget.
Conclusion
The rapid monetary policy tightening around 2022, combined with historically high balance sheets, had a negative effect on the finances of the monitored central banks. Most of them made a loss and those banks that had previously sent transfers to the central government budget from their profits had to suspend them. Several central banks are now operating with negative equity. However, this does not constitute an obstacle to their conduct of monetary policy, as highlighted by these banks, and they will gradually settle these accumulated losses from their future profits. As regards the size of balance sheets, they have been gradually decreasing in most of the monitored central banks over the last two years, as a percentage of GDP and in absolute terms. In the case of many central banks, this is being fostered by ongoing quantitative tightening. In several cases, foreign exchange interventions played a role in strengthening their own currency. From a historical perspective, however, balance sheets are still high. Nevertheless, both the size of the central banks’ balance sheets and the results are not an objective, but a side-effect of the activities of central banks seeking to fulfil their statutory mandates.
References
BIS: “Long series on central bank total assets”, data set documentation (last updated 20 August 2024). (external link)
Bell, S., Chui, M, Gomes, T., Moser-Boehm, P. & Pierres Tejada, A. (2023): “Why are central banks reporting losses? Does it matter?”, BIS Bulletin No. 68, Bank for International Settlements. (external link)
Chaboud, A. & Leahy, M. (2013): “Foreign Central Bank Remittance Practices”, The Federal Reserve Board, Division on International Finance. (external link)
Hall, R. & Reis, R. (2015): “Maintaining Central-Bank Financial Stability under New-Style Central Banking”, NBER Working Paper 21173. (external link)
Long, J. & Fisher, P. (2024): “Central bank profit distribution and recapitalisation”, Staff Working paper No. 1069, Bank of England. (external link)
Rule, G. (2015): “Understanding the central bank balance sheet”, Centre for Central Banking Studies, Bank of England. (external link)
Central banks’ websites and annual reports.
[1] The topic of central banks’ balance sheets is quite broad and covering all the relevant aspects is beyond this article; related issues such as the detailed structure of central banks’ balance sheets and the specific course of quantitative easing and tightening in recent years are not discussed in detail here. Individual quantitative easing programmes and the subsequent turn to quantitative tightening, as well as the transmission of these steps, was described, for example, in the Central Bank Monitoring Spotlight in March 2022. The structure of central banks’ balance sheets is described, for example, by Rule (2015).
[2] For a more detailed discussion of different accounting standards across central banks – as well as differences across countries in sending the central bank’s profits to the central government’s budget – see Chaboud and Leahy (2013).
[3] In the case of converging economies with high foreign exchange reserves on the central bank’s balance sheet, convergence will occur towards more advanced countries if the economy thrives and the central bank succeeds in maintaining price stability. The nominal exchange rate can be expected to appreciate and hence have a negative impact on the central bank’s profitability. By contrast, any poor economic situation leading to a weakening of the domestic currency would lead to higher profits for the central bank, but it would not be positive from the viewpoint of the economy as a whole.
[4] Specific examples of central banks are discussed in the following section.
[5] In the remainder of this article, we focus on central banks regularly monitored in Central Bank Monitoring. For ease of reference, we follow the breakdown of these banks used in the first section of this publication. For comparison, this standard selection of central banks is supplemented by the Bank of Japan, which has long been among the central banks with the largest balance sheets relative to GDP in the world.
[6] Moreover, this historical dataset ends in 2019 and therefore does not include a further increase in the BoE’s balance sheet of over 45% of GDP after 2020 – see Chart 3. The values in the two charts differ slightly from each other due to methodological differences (although they are roughly in line).
[7] In the case of banks where foreign exchange reserves play a significant role, the exchange rate at which foreign currency assets are converted into the domestic currency also plays a role – see the discussion above.
[8] For an overview of the institutional set-up of transfers between central banks and the governments of 70 central banks, including individual euro area Member States, see Long and Fisher (2024).
[9] Although the basic principles of this agreement were known from the outset, the exact wording of the agreement was not public until recently. However, following a decision by the UK authority handling data protection and the right to information, the Treasury had to publish the text of the agreement with the Bank of England, which is available here, in November this year (although some sensitive information was not disclosed).
[10] Between 2009 and September 2022, the combined effect was around GBP 124 bn.
[11] The SNB sent a total of CHF 6 bn to the federal government and cantons in 2020 and again in 2021, thereby providing around 3 % of their income. The SNB’s loss was around CHF 132 bn in 2022 (i.e. several times more than the previous record-high loss of CZK 23 bn in 2015), in 2023 it was about CHF 3 bn. See also this speech by the then Vice-Chairman (current Chairman) of the SNB Governing Board, Martin Schlegel.
[12] The current act sets three limits (which are indexed for inflation) for the value of Riksbank’s capital – a target value (now SEK 62.6 bn), a baseline (two-thirds of the target value) and a minimum (one third of the target value). If it falls below the minimum, the Riksbank must request recapitalisation. The Riksbank asked for a capital injection of more than SEK 40 bn in April, which would have brought its equity up to the baseline value, and finally received SEK 25 bn in September, which brought slightly negative equity just above the minimum. In September, the government also proposed an amendment to the act, which, among other things, would mean the Riksbank had the option of requesting recapitalisation, but did not have to do so.