Dr Sushil Wadhwani, CBE, is currently a Research Associate at the London School of Economics
The independence of the Bank of England (BoE) is perhaps the best thing that has happened to British macroeconomic policy in my lifetime. In recent years, though, parliamentary committees have been especially interested in quantitative easing (QE) (eg. House of Lords 2021, House of Commons 2024) and, a few weeks before being appointed as the First Secretary of State, Louise Haigh, argued that the time was right to re-examine the BOE’s mandate (Haigh 2026). The reforms that I propose below should help deal with some of this political criticism and thereby entrench independence.
What might we do regarding the losses associated with QE
The Office for Budgetary Responsibility’s most recent published estimates suggest that the cumulative net lifetime loss from the BOE’s Asset Purchase Facility (APF) might amount to £164 billion by the end of 2036 (OBR 2025, para 6.18).
While the OBR emphasises that this excludes the wider benefits of the QE programme, it nevertheless shows that a government that is committed to its fiscal rules ends up probably having to spend less or tax more than might have been true in the absence of these losses.
Recall that the procedures associated with QE were put together at a time of great stress for all involved and it would be reasonable to expect that these can be enhanced. I therefore make five recommendations below with regard to how we might reduce these APF losses in the future, but it is best to understand the underlying problem first.
Consider the stance of the Monetary Policy Committee (MPC) that it pays no attention whatsoever to the profit or loss associated with the assets transacted when making their monetary policy decisions. This is because the government indemnifies the BOE against future losses. One justification offered by the BOE is that when it buys a gilt, its expected return on the overall transaction (which typically involves financing the gilt purchase through paying interest on reserves) is approximately zero because markets are efficient and accord with the expectations theory of interest rates1.
In this regard, Shiller (1979) showed that long-term bond yields move around substantially more than the conventional expectations theory of the yield curve can justify. An implication is that bond returns are somewhat predictable and so, when buying a longer-term gilt, there are some times when you want to be a buyer and others when you wish to be a seller.
It is important to remember that central banks have typically launched QE in difficult situations involving a financial crisis. Think of them as periods when ‘animal spirits’ are unusually low, as perhaps evidenced by stock prices being ‘irrationally’ low too. Investors typically flee to the safe haven provided by longer-term bonds and this depresses the term premium.
So, central banks know that if and when the economy normalises, interest rates and the term premium will both rise and bond prices will fall. It is therefore likely that QE will ultimately lose money as it involves ‘buying high, selling low’.
The first recommendation is that, given that accountability is important, the MPC should revisit its efficient markets view and publish scenarios contemporaneously showing that significant APF losses might materialise and why the Committee nevertheless believes that the QE operations are still likely to deliver a net benefit to the economy as a whole. The very act of having to be explicit in this regard should help focus minds on the MPC and exert downward pressure on losses.
The BOE also justifies its stance on ignoring the expected losses associated with QE by saying it needs to exclusively focus on achieving the inflation target and not be distracted by fiscal policy considerations. But the MPC can achieve the inflation target in multiple ways, and differences in the level of expected losses can help them choose between these alternative strategies.
For example, the MPC could choose to buy UK equities instead of gilts during these periods of panic in order to help achieve its inflation target (the Bank of Japan also bought stocks in recent years). An advantage of doing so is that buying equities would have made money.
The two suggestions made to enhance the credibility of the BoE and the inflation targeting regime should both help reduce the debt interest bill and the cost of reducing inflation. The recommendations with regard to reducing QE losses should hopefully make central banking more boring and reduce the degree of unwanted attention from politicians
So the second recommendation is that the MPC should discuss and show comparisons in likely costs between alternative ways of achieving the inflation target (eg. buying equities versus gilts, buying gilts at different points along the curve, etc). I might note here that Australia, in implementing the suggestions of a review of its central bank, has agreed that the Monetary Policy Board will include the benefits, costs, and risks associated with available tools when contemplating their use2.
A third recommendation is to facilitate greater policy coordination between the BOE and the Treasury. The MPC would, initially privately, share its analysis of its particular form of proposed QE stimulus, showing the expected fiscal costs in comparison to the expected benefits (as per the recommendations above). The Treasury would then be given the option to propose an alternative fiscal policy measure which provides a superior benefit/cost ratio while also helping to deliver the inflation target. Minutes of these discussions would be published with a time lag.
An exception would need to be made for situations where the MPC believed that action on its part was very urgent. Of course, the details of such an arrangement deserve further consideration as while it would be desirable in terms of ensuring that the return for a given fiscal cost is optimised, it must be compatible with the BOE being seen to be independent.
A fourth recommendation is that the MPC should revisit whether the QE instrument has become unnecessarily inflexible because of the Committee’s tendency to pre-commit to purchases over a long period. For example, the MPC committed itself on 4 November 2020 to buy £150 billion of gilts over the subsequent year, and yet on 9 November, highly encouraging vaccination results were announced.
If, as per earlier suggestions, the MPC had been told to consider the expected profits/losses associated with QE in conjunction with the overall economic benefit, it would probably have truncated the pre-announced QE programme after the vaccine news.
A fifth recommendation is to reduce the fiscal costs of QE by operating a tiered reserves remuneration structure (as the ECB does; see De Grauwe and Ji 2024). Of course, the details of the change here should pay due consideration to the need to preserve monetary policy transmission.
How Bank of England reform might help strengthen credibility
I have previously discussed on VoxEU that the level of breakeven inflation embedded in UK gilt yields was high in relation to its inflation target and international peers (Wadhwani 2024a). This remains true today3. The 20-year breakeven inflation rate in the UK is currently around 3.32%, which is high in relation to the inflation target of 2%.
We do though need to account for a quirk whereby UK index-linked gilts compensate investors for Retail Prices Index (RPI) inflation for the first four of these twenty years, and Consumer Price Index including occupiers’ housing costs (CPIH) inflation from 2030 onwards. Note that the four-year ahead breakeven inflation rate is approximately 3.85%.
This then implies that the expected average inflation compensation for the 16-year period beginning in four years that is embedded in markets is around 3.19%, which is well above the inflation target of 2% and is broadly in line with the level of 3.17% reported in Wadhwani (2024a).
By comparison, the level of breakeven inflation in the US is around 2.40% (for CPI inflation), which is broadly in line with the Federal Reserve’s target of 2% for Personal Consumption Expenditures (PCE) inflation. There is therefore a significant ‘inflation risk premium’ embedded within UK gilts, while investors in US Treasuries do not appear to need such a risk premium.
One possible reason for the higher inflation risk premium in the UK is that the government sets the inflation target while, in the US, the Federal Reserve chooses its own definition of price stability (ie. goal independence). I am tempted to recommend that the BOE be allowed to set its own target, thereby bringing it in line with the US and the ECB.
But some have argued that the operational independence of the BOE was only feasible because Parliament retained control of the inflation target (Stansbury and Balls 2017). Therefore, an alternative way forward is to enshrine the inflation target into law using primary legislation, as this would still signal that the government is serious about entrenching price stability.
Another possible reason as to why the inflation risk premium is elevated in the UK is that the BOE itself is perceived as less credible than some other central banks. The proposals within the Bernanke Review (Bernanke 2024) were regarded as a way to help enhance BOE credibility, but as I argued at the time (Wadhwani 2024b), the forecasting errors made by the Bank had much more to do with a lack of intellectual diversity.
By way of example, the Bank had failed, at the end of 2020, to notice that its forecasts were incompatible with the ongoing money growth and most reasonable beliefs regarding the behaviour of velocity. It therefore does not surprise me that the level of the inflation risk premium has remained broadly stable in recent years even after the implementation of some of the recommendations of the Bernanke Review, as the markets appreciate that more needs to change4.
The fact that inflation has spent much of the last five years above target has not helped. Specifically, inflation moved above target in May 2021 and has been above target in all but three months since then (it was at the 2% target in May-June 2024 and below target in September 2024). Therefore, I recommend that the review looks afresh at recruitment procedures to ensure that we get a more intellectually diverse membership.
Haigh (2026) has suggested that the BOE remit be amended to pay more attention to growth by, perhaps, switching to a dual mandate. Recall that the BOE’s statutory objective is to maintain price stability and, subject to that, support the government’s economic policy (including growth). Moreover, the remit explicitly allows the MPC to return inflation to target gradually when immediate disinflation would generate excessive output losses.
So, I am yet to be convinced that switching to a dual mandate would make much difference in practice to how the MPC actually sets interest rates. On the other hand, there is a risk that it would be misunderstood by markets as a signal that the government is less committed to the inflation target and might thereby widen the inflation risk premium.
Conclusion
The two suggestions made to enhance the credibility of the BoE and the inflation targeting regime should both help reduce the debt interest bill and the cost of reducing inflation. The recommendations with regard to reducing QE losses should hopefully make central banking more boring and reduce the degree of unwanted attention from politicians.
Endnotes
1. See, for example, Ben Broadbent’s response.
2. See Mahon and Milas (2026) for further discussion.
3. Using data on 10 September, 2026.
4. On the day before the publication of the Bernanke Review (11 April 2024), the estimated six-year-ahead, fourteen-year breakeven inflation rate was 3.23%, which compares to our estimate of underlying long-term inflation expectations of 3.19% now.
References
Bernanke, B (2024), Forecasting for monetary policy making and communication at the Bank of England: a review, Bank of England.
De Grauwe, P and Y Ji (2024), “The new operating procedures of the Bank of England”, VoxEU.org, June 24.
Haigh, L (2026), “A new fiscal framework to renew Britain”, Renewal 34, May.
House of Lords (2021), “Quantitative easing: a dangerous addiction?”, July 16.
House of Commons (2024), Quantitative Tightening, fifth report, January 31.
Mahon, C and C Milas (2026) “Lessons in letter writing for the Bank of England”, Alphaville, Financial Times, 10 August.
OBR – Office for Budgetary Responsibility (2025), Economic and fiscal outlook, November 2025.
Shiller, R (1979), “The Volatility of Long-Term Interest Rates and Expectations Models of the Term Structure”, Journal of Political Economy 87(6): 1190-1219.
Stansbury, A and E Balls (2017), “Twenty years on: Is there still a case for Bank of England independence?”, VoxEU.org, 1 May.
Wadhwani, S (2024a), “How to cut the UK debt interest bill”, VoxEU.org, 22 July.
Wadhwani, S (2024b) “The Bernanke Review: A Missed Opportunity?”, in D Aikman and R Barwell (eds), The Bernanke Review: Responses from Bank of England Watchers, King’s Business School.
Editors’ note: The author is a former member of the Monetary Policy Committee at the Bank of England. This article was originally published on VoxEU.org.