UK Macro View: Can the government recoup the fiscal cost of QE?

Click here for full report and disclosures

● Rationale for a tax rise? As the Bank of England (BoE) winds down its quantitative easing (QE) programme, its losses will cost the new government around £27bn a year (equivalent to 1% of GDP) over its five-year term in office. While the major central banks are now making losses on QE, the way this is accounted for in the public finances in the UK is uniquely impactful. Ways to prevent such a large loss for the exchequer include the BoE stopping interest payments on some portion of banks reserves, but the BoE has dismissed the idea. In any case, most of the losses will be due to the falls in the prices of the gilts the BoE purchased. Instead, the UK government could change its accounting rules so that the losses do not force it to tighten fiscal policy. Alternatively, it may use the cost of the scheme as rationale for raising tax on banks in the budget on 30 October.

● Unconventional policy has made the public finances more vulnerable: QE has expanded the BoE’s balance sheet to an unprecedented size and exposed the public finances to greater interest rate risk. From the perspective of the consolidated public sector balance sheet – ie the government and BoE together – QE amounts to a large debt management operation. Long-term government debt is retired and replaced with reserves on which the BoE pays interest at Bank Rate. As the BoE owns almost a third of UK government debt, this tied the government’s debt interest costs more closely to the policy rate, which has proven costly as it has risen. It also made the government responsible for losses on the BoE’s gilt portfolio.

● Fiscal cost will ramp up as the BoE crystallises losses: Until 2021, the BoE made profits because it had replaced higher-yielding gilts with reserves it remunerated at Bank Rate, which was close to zero. However, as it raised the policy rate above the yields on the gilts it purchased, the higher cost of interest on reserves outstripped income from the gilt portfolio. Moreover, by starting quantitative tightening (QT), the BoE has begun to realise major losses on bonds it purchased at much higher prices – see Chart 1. Indeed, the sale and maturity of gilts at a lower price than the BoE bought them for is expected to account for the majority of the losses over the course of QT – see Chart 2.

● Debate over which debt measure to use for the fiscal rule misses the point: The fiscal rules stipulate that the debt-to-GDP ratio must fall in the fifth year of the Office for Budget Responsibility’s forecast. Chancellor Rachel Reeves is reportedly considering changing the metric of national debt used from Public Sector Net Debt (PSND) excluding BoE holdings to the measure that includes them. This change could give the government an extra £17bn of “headroom” to loosen fiscal policy. One reason the tweak would help is because it pushes some of the losses made by the QE programme into the past. Under PSND ex-BoE, the increase in national debt due to QE losses is realised when HM Treasury reimburses the BoE. However, in PSND, the difference between the (higher) purchase price the BoE paid for the bond and par value is booked when the bond was bought. Ultimately, both measures will leave the national debt at around 95% of GDP, of which losses made by the QE programme will account for c4ppt.

● International comparisons: Including QE losses in measures of the government’s deficit and national debt in real time is unusual. The Federal Reserve (Fed) and even the Bundesbank keep losses on the balance sheet as a deferred liability, to be paid down over time with seigniorage revenues. Even so, the ECB has stopped paying interest on banks’ minimum required reserves to curb its losses, thereby improving the “efficiency” of monetary policy. BoE governor Andrew Bailey says that the UK central bank can not run down the loss from QE over time like the Fed because it gives its seigniorage revenue to HM Treasury. Furthermore, it does not have a minimum reserve requirement like the ECB as a “matter of history”.

● Policy solutions: As shown above, non-payment of interest on some bank reserves would not temper the majority of losses the BoE will make over the remainder of the QE programme. However, the impact on the public finances of the current arrangement suggests that it will come under scrutiny in the budget. That could lead to a change in accounting practices, but the risk is that it is also used as a rationale to increase taxation on banks. Indeed, QE has raised banks’ interest income, absorbed a significant share of the losses from falls in gilt prices and prevented a worse economic outcome at a increasingly significant cost to the government. It can be argued that QE itself exaggerated the rise in interest rates and corresponding fall in bond prices that is causing the government’s losses. Even so, the government could well use the fiscal cost of QE to justify an increase in taxation on the institutions that have benefited most directly from it. Therefore, some increase in the corporation tax surcharge on bank profits seems likely in the budget.

Andrew Wishart

Senior UK Economist

[email protected]

+44 20 3753 3017

Disclosures

This material is intended as commentary on political, economic or market conditions for institutional investors or market professionals only and does not constitute a financial analysis or a research report as defined by applicable regulation. See the "Disclaimers" section of this report.

The commentary included herein was produced by Joh. Berenberg, Gossler & Co. KG (Berenberg). For sales inquiries, please contact:

Phone: +44 (0)20 3207 7800
Email: [email protected]

BERENBERG
Joh. Berenberg, Gossler & Co. KG
Neuer Jungfernstieg 20
20354 Hamburg
Germany

Registered Office: Hamburg, Germany
Local Court Hamburg HRA 42659

For Berenberg the protection of your data has always been a top priority. Please find information on the processing of personal data here.
Any e-mail message (including any attachment) sent by Berenberg, any of its subsidiaries or any of their employees is strictly confidential and may contain information that is privileged or exempt from disclosure under applicable law. If you have received such message(s) by mistake please notify the sender by return e-mail. We ask you to delete that message (including any attachments) thereafter from your system. Any unauthorised use or dissemination of that message in whole or in part (including any attachment) is strictly prohibited. Please also note that any legally binding representation needs to be signed by two authorised signatories. Therefore we do not send legally binding representations via e-mail. Furthermore we do not accept any legally binding representation and/or instruction(s) via e-mail. In the event of any technical difficulty with any e-mails received from us, please contact the sender or [email protected]. Deutscher disclaimer.

Click here to unsubscribe from these emails.