QT0006
Written evidence submitted by Positive Money
Positive Money welcomes the opportunity to respond to the Committee’s inquiry into quantitative tightening.
We are a not-for-profit research and campaigning organisation, working towards reform of the money and banking system to support a fair, democratic and sustainable economy. We are funded by charitable trusts, foundations and small donations.
Our submission makes the following key points:
● The Bank of England does not appear to have sufficiently understood or prepared for the impact of quantitative tightening (QT), as illustrated by the events surrounding the ‘mini-Budget’.
● The Bank of England has not devised an appropriate strategy for QT (and monetary tightening more generally) that reduces the cost to the public. The Bank’s insistence on active gilt sales represents particularly bad value for the public, and it is unclear what the Bank’s rationale for this is.
● These costs to the public result from the way in which HMT has indemnified the Bank of England against any losses, which in itself raises the spectre of moral hazard. At the very least the UK should follow the approach of the US, where rather than requiring the central bank to be ‘bailed out’ against losses on asset purchases, losses are simply recorded as a deferred asset until they can be recouped.
● There are a number of further ways in which policymakers can reduce the significant fiscal costs of monetary tightening, including changes to reserve remuneration, altering the maturity structure of the public sector’s liabilities, as well as more direct windfall taxes on the beneficiaries. There are both international and historical examples the Bank and HMT could learn from.
● Just as quantitative easing (QE) itself did little to increase consumer price inflation, it is unlikely QT will do much to reduce it.
● Though QE has played a significant role in increasing wealth inequality, it is unlikely that QT will play a positive role in undoing this.
- Have the Bank of England and the Monetary Policy Committee developed an appropriate strategy and framework for quantitative tightening? Are there any successful international or historical examples to follow?
1.1. We are concerned that the Bank of England has not devised an appropriate strategy and framework for quantitative tightening (and monetary tightening more generally) that reduces the cost to the public. There are both international and historical examples the Bank and HMT could learn from.
1.2. The Bank of England’s current approach to QT, if followed through, would cost the government more than £230bn over the next decade, according to the most recent analysis the Bank has published, greatly outweighing any fiscal benefits so-far accrued from QE.[1] These costs result from not only from reserve remuneration (as discussed further in 4.3), but also the Bank of England’s decision to undertake ‘active QT’ and sell bonds while prices are much lower than the time of purchase, crystallising mark-to-market losses.
1.3. The significant net losses expected from the Bank of England’s gilt sales represent unjustifiably bad value for public money given that HM Treasury indemnifies the Asset Purchase Facility against any losses, ultimately leaving the public liable. It is particularly concerning that the Bank and Treasury have failed to make the full Deed of Indemnity public, with the Chancellor even blocking requests for more transparency from the House of Lords’ Economic Affairs Committee.[2]
1.4. We are concerned that the Bank of England may be undertaking aggressive QT in order to combat accusations of ‘fiscal dominance’, with the knowledge that it is not the Bank but the Treasury who will be liable for losses. Akin to the moral hazard of too big to fail banking, this creates a perverse structure of incentives where Bank of England policymakers may be minded to furnish their reputations for ‘credibility’ using, in the parlance of Adam Smith, “other people’s money”[3] — in this case the public’s money. Unless the members of the MPC are willing to risk their own capital, this seems a wholly inappropriate set-up.
1.5. It is unclear what is steering the Bank of England’s decision to time bond sales in a way which would crystallise such significant losses for the public, especially given that Governor Andrew Bailey has said that “the MPC is not using the stock of asset holdings as an active tool of monetary policy at present.”[4] This has been echoed by the Bank’s Executive Director for Markets, Andrew Hauser, who reaffirmed that QE is not an “active tool for monetary policy tightening.”[5]
1.6. We also note that of the four major central banks — the Federal Reserve, European Central Bank, Bank of England and Bank of Japan — the Bank of England is the only one undertaking ‘active’ QT by selling bonds to the market before maturity. We see no clear justification for this divergence from the Bank’s international counterparts.
1.7. Regardless of whether the Bank of England is right to sell bonds at this time, HMT using public money to ‘bailout’ the Bank from the resulting losses is unnecessary, as the experience of other central banks illustrates. Unlike commercial banks, central banks are able to operate with negative equity,[6] as the likes of Switzerland and Sweden have shown. Moreover, the UK could follow the same approach as in the United States, where the Treasury benefits from the profits from asset purchases, but is protected against losses, which are instead recorded as deferred assets until they can be recouped.[7] There appears no clear reason why the Bank of England can’t do the same, instead of the Treasury having to cover losses unnecessarily, at significant cost to the public.
1.8. While not precisely equivalent to QT, there may also be historical precedent in the way the Bank of England implemented tighter monetary policy in the early 1950s, when, similar to today, the government had a large stock of debt that was sensitive to a rise in short-term interest rates. When it started raising interest rates in 1951, the Bank and Treasury coordinated to minimise undesirable side-effects, through measures such as the Serial Funding operation. Policymakers were able to reduce the costs to the public purse by requiring financial firms to exchange their short-term Treasury bills for longer-maturity Serial Funding stocks, locking in lower interest rates. Writing for the National Institute of Economic and Social Research (NIESR), William Allen, Jaghit Chadha and Philip Turner have put forward proposals for a similar approach today, in which central bank reserves could be exchanged for special issues of fixed-interest gilts over a range of maturities.[8] This is one way in which policymakers could mitigate the fiscal impact of monetary tightening today, other options are outlined in section 5.
- What will be the impact of quantitative tightening on inflation, the economy, households, the gilt market, and the wider financial sector? Are these impacts and any risks around them well understood?
2.1. The Bank of England’s Independent Evaluation Office recently concluded that the Bank has a poor understanding of how QE works as well as its effects.[9] It is therefore likely that the Bank also lacks a sufficient understanding of its reverse, QT. This is particularly worrying given that the Bank of England is taking a more aggressive approach to QT than its peers. As noted above, while the Fed and ECB have been taking a more ‘passive’ approach to tightening (first by slowing the rate of purchases and second by not reinvesting the proceeds from maturing bonds), the Bank of England is the only major central bank that has sought to actively sell its holdings of gilts.
2.2. The events surrounding the ‘mini-Budget’ of Autumn 2022 may indicate that the Bank was unprepared for or did not fully understand the risks of its approach to QT. Yields on 30 year gilts started climbing in August after the Bank of England released a market notice saying it was provisionally minded to commence gilt sales from September.[10] Despite a relatively dovish[11] 50bps increase to Bank Rate, yields rose further steeply when the MPC confirmed plans to begin selling gilts shortly in its meeting on September 22, the day before the ‘mini-Budget’.[12] Though it is difficult to isolate the effects of QT, this indicates that it may have been a driving factor in long-dated gilt yields rising to levels that threatened financial stability and forced the Bank to undertake new emergency asset purchases. The Bank of England’s U-turn on QT quickly reversed the fall in bond prices/rise in yields. Regardless, it appears the Bank of England may have been overly complacent about the risks of its policy, or was blind to the scale of leverage among the pension funds.[13] We note that in 2017 the Bank’s Deputy Governor for Financial Stability Sir Jon Cunliffe claimed that “Pension funds are active in repo markets but do not run highly leveraged positions”[14] — it is unclear what changed between then and September 2022.
- Was there an impact from quantitative easing on inequality and will there be any impact of quantitative tightening on inequality?
3.1. QE has played a significant role in exacerbating wealth inequality through its effect on asset price inflation and the highly unequal distribution of assets. The Bank of England’s own analysis suggests that the richest 10% of households saw their wealth increase by an average of £350,000 between 2006-08 to 2012-14 as a result of QE — more than 116 times the benefit seen by the poorest 10%.[15] This figure has likely grown much higher following further rounds of QE, and we can expect the gains to be even more unequal for the very top percentiles.
3.2. However it is unlikely that QT will have a positive impact on redressing this inequality. As the reverse of QE, QT could reduce the wealth of the asset rich, but it won’t increase the wealth of the asset poor. On the contrary, tighter monetary policy most negatively affects ordinary workers who face unemployment and stagnating wages, while the cash rich could see greater returns from higher interest rates.
- What are the fiscal impacts of quantitative easing and tightening? What ways might there be of reducing the fiscal costs forecast to be incurred over the next few years, and what would be the benefits and costs of doing so? What approach are the US Federal Reserve and the European Central Bank taking to the fiscal impact of quantitative tightening and how and why do they differ from that being taken by the Bank of England?
4.1. In November 2022 HMT made a first transfer of £11bn to the Bank of England to cover losses on the Asset Purchase Facility.[16] As discussed in 1.2, the Bank of England’s current approach to monetary tightening will, if followed through, cost the government more than £230bn over the next decade, according to the most recent analysis the Bank has published.[17] This would greatly outweigh any fiscal benefits accrued from QE and see the state taking a significant net-loss, representing unjustifiably bad value for money for the public.
4.2. Through changing the composition of the non-state sectors’ holdings of public liabilities from gilts to central bank reserves, QE has made the government’s debt servicing costs much more sensitive to changes in short-term interest rates. This is because central bank reserves are currently remunerated at the Bank of England’s base rate, whereas rates are fixed for a longer duration for other government liabilities such as gilts.
4.3. As the Bank of England raises its base rate, the amount paid remunerating reserves exceeds the income received from the bonds the Bank acquired when interest rates were lower. Alongside the capital losses discussed in 1.2, any shortfall in this interest flow is also covered by HMT, which indemnifies the Asset Purchase Facility, impacting the government’s fiscal position. The fiscal cost of reserve remuneration will be significant as the Bank raises interest rates, with the New Economics Foundation estimating that the central bank is set to pay banks around £150bn in interest on their risk-free reserves between 2022 and 2028. This estimate is consistent with the figures published by the OBR in November 2022.[18]
Notes: Colours of bars refer to marginal difference. In all scenarios active sale of gilts by the asset purchase facility continues until November 2023 while the roll off of gilts at maturity continues throughout the period. In the ‘stops’ scenario active sales are assumed to stop in November 2023, in the ‘continues’ scenario the sales carry on at the same pace (£45bn over 14 months) and in the ‘speeds up’ scenario the pace doubles (£90bn over 14 months). Interest rate expectations as of end of February 2023.
Source: New Economics Foundation analysis of Bank of England data.
4.4. As noted in 1.6, the Federal Reserve is taking a different approach to monetary tightening to reduce the costs to the public, by recording losses as a deferred asset. Like the Federal Reserve, the ECB is holding government bonds to maturity, rather than selling at a loss like the Bank of England, though there are still potential fiscal costs from reserve remuneration. The ECB however has already introduced tiered reserve remuneration as discussed below, which could be used to reduce costs.
- What lessons should be drawn for the design and operation of any future rounds of quantitative easing and tightening? In particular, are there any suitable ways of reducing the fiscal impacts of any future rounds?
5.1. Firstly, a key lesson since the introduction of QE in 2009 is that monetary policy should not be the principal means of managing demand. The main mechanism through which QE stimulates the economy is through relatively weak ‘trickle down’ wealth effects, which stem from increased demand for assets as those who have sold gilts to the Bank of England reinvest the proceeds in order to rebalance their portfolios.[19] The Bank of England’s own analysis suggests that for every £1 of QE, only 8p fed into the real economy.[20] A much smaller fiscal stimulus would likely have provided the same support to the economy with less of the negative side effects, such as asset price inflation.
5.2. As QE works by encouraging riskier investments through the search for yield and increasing private debt, it also stokes financial fragility. This makes it difficult to respond to an ‘overheating’ economy with sudden monetary tightening without risking financial stability, as appeared to be the case with the LDI crisis in September 2022, creating a clear tension between the Bank of England’s objectives for price and financial stability. At the time of writing, we are also seeing monetary tightening reduce the mark-to-market value of banks’ assets in a manner which threatens their solvency, most notably in the case of Silicon Valley Bank.[21]
5.3. Asset purchases like QE can perhaps more usefully be thought of as financial stability tools. The Bank of England’s bond purchases were effective in providing liquidity to distressed government debt markets and preventing a dangerous fall in bond prices during the COVID-19 pandemic, and subsequently stabilising long-dated gilt markets in September 2022.
5.4. Where asset purchases are deemed necessary, there are a number of ways in which they could be designed and implemented more optimally. In the interests of fiscal prudence, expansionary monetary policy could be managed in ways that ensure the public sector is less exposed to rising short-term interest rates. One such way would be to convert the public liabilities introduced by asset purchases from reserves to securities of longer duration. The opportune time to do this was before interest rates started rising to lock in lower rates, and it is unclear why it appears policymakers didn’t consider this if they were expecting a higher Bank Rate. Nevertheless, as referred to in 1.7, if the Bank is still intent on tightening policy it may still be possible to reduce costs by exchanging variable-rate reserves for special issues of fixed-interest gilts over a range of maturities.
5.5. Another option is to return to a system of non-remunerated reserves, as existed before the introduction of QE in 2009. This would require introducing alternative means of implementing monetary policy if banks still had excess reserves (the interbank lending rate would fall to zero otherwise), such as the reintroduction of reserve requirements. Instead of paying banks to hold reserves, demand for reserves in an ample reserve environment could also be maintained through an unreserved deposit tax. Levying a tax on deposits not backed by reserves equal to the policy rate would enable the implementation of monetary policy with the added benefits to the public of increasing the safety and soundness of banks and recapturing seigniorage. For instance, a tax on all unreserved deposits at the current 4% Bank Rate would raise more than £50bn a year. As with reserve remuneration, this could also be implemented on a tiered basis, as discussed below.
5.6. Alternatively, the Bank of England could introduce a ‘tiered’ system of reserve remuneration, as other central banks such as the ECB, Swiss National Bank and Bank of Japan have introduced. Rather than remunerating banks’ entire holdings of reserves, as is currently the case, a tiered system would enable the Bank of England to restrict the payment of full Bank Rate to a smaller amount of reserves as necessary for the implementation of monetary policy, with lower (or zero) rates paid on the rest. The New Economics Foundation estimated in June 2022 that moving to a tiered reserve system could save the government between £25bn and £57bn in income transfers to the banking sector by March 2025.[22]
5.7. As reducing reserve remuneration can be considered an implicit tax on the banking sector (or, more accurately in our view, the removal of a subsidy), it may be more desirable to levy a more explicit windfall tax on banks’ unearned profits from higher interest rates, as the government did in 1981 (alongside windfall taxes on energy companies).[23] However, the current government appears to be moving in the opposite direction. Despite the introduction of a 35% Energy Profits Levy on oil and gas companies’ windfalls, the government has at the same time cut the existing surcharge on bank profits by 60%, from 8% to 3%. A windfall tax on banks, as is being considered in other countries such as Spain,[24] would be a simple and equitable way of offsetting the fiscal impact of monetary tightening banks are profiting from.[25]
- What role did quantitative easing, its timing and its interaction with wider economic policy play in the outbreak of double-digit inflation? What effects will quantitative tightening and its timing have on inflation and growth?
6.1. QE itself is unlikely to have had a significant role in the outbreak of double-digit inflation, just as it failed to have a significant impact on inflation in the preceding decade. Fears of QE causing excessive inflation are usually based on a commonly-held but incorrect ‘money multiplier’ view of banking,[26] and misunderstandings of the function of central bank reserves, including a mistaken belief that banks are able to ‘lend out’ the reserves created by QE. Central bank reserves are a form of money used to settle transactions between banks and with the state, as well as for the implementation of monetary policy. Reserves can only be held in accounts at the central bank and the only way they can feasibly ‘leak’ into the economy is if banks initiate new lending and customers decide to exchange the deposits created for much greater quantities of physical cash (a relatively unlikely prospect).
6.2. Furthermore, despite what is inferred from the common conception of QE as ‘money printing’, QE does not simply involve the central bank freely dispersing money into the economy. In actuality, QE money is exchanged for government bonds. Government bonds are extremely ‘money-like’, as highly-liquid assets that can be easily exchanged for cash on demand and that also play a central role as collateral, making them key instruments for credit generation particularly in the shadow banking sector. QE is therefore best understood as an asset swap - while it increases the private sector’s holding of one highly-liquid asset, it equally decreases holdings of another. As such, QE itself cannot be considered to have increased net-purchasing power in the real economy, and so is unlikely to have caused significant increases in consumer prices.
6.3. However, the effect of QE on inflating asset prices is clearer. The Bank of England’s own analysis suggests that real house and share prices in 2014 would have been 25% and 22% lower respectively in the absence of monetary easing.[27]
6.4. It can more feasibly be argued that the fiscal expansions pandemic QE accompanied (and perhaps enabled) may have had a greater role in increasing demand and inflationary pressure. While QE simply changes the composition of the non-government sectors’ assets, the government’s deficit spending directly increases net-financial assets and net-purchasing power. However the absence of such government spending would have been even more disastrous, and unwanted side-effects could have been mitigated by raising taxes, particularly on those whose wealth had increased significantly during the pandemic.[28]
6.5. Regardless, the chief drivers of the UK’s inflation are clearly supply-side, namely the effect of the pandemic on global supply chains and huge increases in the prices of energy and food stemming from the war in Ukraine. The Bank of England has said these non-domestic factors accounted for over 70% of inflation in 2022. Domestic factors — profit margins, labour costs and, to a lesser extent, taxes — accounted for the remaining 30%, yet according to the Bank this may even be an overestimate of domestically generated inflation.[29]
Notes: ‘GOS’ refers to profits and ‘UWCs’ refers to total unit labour costs.
Source: Bank of England[30]
6.6. Just as QE itself has had little impact on increasing consumer price inflation, it is also unlikely that QT will have much effect on reducing it, particularly as inflation has been driven by supply-side factors.
March 2023