How much central banks might lose on their engorged bond portfolios as interest rates rise and whether these losses even matter has been a subject of interest at FTAV Towers for a while. Now the BIS has also tackled it.
It’s a hot topic, given the scale of the already realised losses, what is to come, and (unfortunately) the political optics in some countries. Central bank accounting is pretty esoteric stuff, but that doesn’t mean that some politicians might not try to weaponise it.
The Bank for International Settlement’s general manager Agustín Carstens has written up his thoughts on the matter for the FT today, but FT Alphaville wanted to dive into the underlying paper itself.
So far ~deep breath~ the Reserve Bank of Australia, the National Bank of Belgium, the Bank of England, the Bank of Japan, the Netherlands Bank, the Swiss National Bank, the Czech National Bank, the Reserve Bank of New Zealand, Sveriges Riksbank and the US Federal Reserve have all already announced that they are facing losses on their asset purchases, according to the BIS report.
Notably, the BIS has in the past been on the sceptical side when it comes to quantitative easing. But the report — authored by Sarah Bell, Michael Chui, Tamara Gomes, Paul Moser-Boehm and Albert Pierres Tejada — is clearly in the “nothing to see here, move along” camp.
You can find the full report here, and here are their main points:
— Rising interest rates are reducing profits or even leading to losses at some central banks, especially those that purchased domestic currency assets for macroeconomic and financial stability objectives.
— Losses and negative equity do not directly affect the ability of central banks to operate effectively.
— In normal times and in crises, central banks should be judged on whether they fulfil their mandates.
— Central banks can underscore their continued ability to achieve policy objectives by clearly explaining the reasons for losses and highlighting the overall benefits of their policy measures.
Basically, the BIS is saying that the losses don’t matter, they don’t affect a central bank’s ability to operate, should be seen in a wider context and should just be explained a bit better.
The heart of the issue is that normal concepts of accounting and solvency don’t really apply when it comes to central banks, which can, well, create money, and are just one side of the overall sovereign balance sheet.
Diving a bit deeper though, central banks take different approaches to how their profits and losses get tallied and reported. Here is a good overview:
From the BIS report:
The three main accounting approaches (Part A in Table 1) affect the size and volatility of net income from asset valuations in the short term, although the results wash out over the longer term. 6 For central banks that use fair value accounting, eg the RBA and BoE, current losses from declines in asset valuation have been front-loaded, and future valuation gains will be reflected as revenue as the assets approach maturity. Others, eg the Eurosystem and Sveriges Riksbank, reflect declines in asset values in reported losses, but reflect unrealised gains only in revaluation accounts. For those that use historic cost accounting, eg the Federal Reserve, unrealised valuation changes are disclosed for transparency, but not recognised in reported income.
Income recognition and distribution rules (Part B in Table 1) determine the size of buffers against losses. These vary considerably across central banks. Some can establish discretionary loss-absorbing buffers before accounting P&L is calculated (eg NBB and DNB). Some make the size of the profit distribution contingent on targets for various types of buffers (eg the Riksbank and BoE). Some also use distribution-smoothing mechanisms, such as distributions based on rolling averages, to make profit transfers to government more predictable over a longer horizon. While these arrangements may reduce transfer volatility and offset accounting losses, they are unlikely to be sufficient to do so under all circumstances.
Indemnity arrangements (Table 1, Part C) may reflect a desire to insulate the central bank from the financial consequences of some policy measures. For example, the BoE APF, established as a subsidiary to conduct APPs, is fully indemnified by the UK Treasury. 7 In other cases (eg RBNZ), the government authorised indemnities for specific operations without a subsidiary. In contrast, some central banks such as the RBA, do not have indemnities. Central banks that have indemnity arrangements view them as a way to ensure that policy measures are not constrained by the prospective financial impact on the central bank, thereby preserving independence. Some that do not have indemnities note that they are irrelevant from the perspective of the overall public sector balance sheet and could even risk reducing policy effectiveness if they weaken perceptions of central bank independence.
However, whatever the approach taken, central banks have no minimum capital requirements, cannot become insolvent in a conventional way, and even sizeable losses don’t compromise a central bank’s ability to operate.
For example, the central banks of Chile, the Czech Republic, Israel and Mexico have all had years of negative capital without impeding their primary job of ensuring financial and price stability, the BIS notes.
The one caveat is when “misperceptions and political economy dynamics can interact with losses to compromise the central bank’s standing”. But for the most past the BIS is relaxed, concluding:
. . . A central bank’s credibility depends on its ability to achieve its mandates. Losses do not jeopardise that ability and are sometimes the price to pay for achieving those aims. To maintain the public’s trust and to preserve central bank legitimacy now and in the long run, stakeholders should appreciate that central banks’ policy mandates come before profits.