Figures out last month showed the quantitative easing (QE) programme is on course to lose £170 billion over its lifetime
- The Bank of England’s (BoE) unwind of its QE programme is radically different to other central banks – making the BoE something of an outlier
- The BoE may have to change course. Questions are being asked. Hints are being dropped. Markets (and taxpayers) would cheer
Figures published last month by the Bank of England (BoE) updated the forecasts of the economics of the QE programme to a projected £170 billion net cost1.
This is a huge figure – about 7.5% of gross domestic product (GDP) and will have to be borne by the Treasury. Other central banks also have QE costs but the BoE unwind of quantitative easing (QE) is generating by far the worst headlines. This is because the BoE QE programme is unique in three ways:
- During the QE era, the BoE bought proportionally much more government debt than other central banks – leaving it most exposed when rates went up.
- Only the BoE is actively selling government bonds which locks in the losses at the current low prices. Other central banks are not actively selling and holding the bonds to maturity, avoiding immediate capital losses and giving time for the interest rate environment to improve.
- While other central banks treat any costs as an accounting entry with little real world implications, the BoE requires any losses to be immediately made good by the taxpayer.
In the world of central banking, these differences make the Bank something of an outlier when it comes to QE. Any attempts to bring it in line with other central banks would be cheered by the market – and the Treasury.
More on that later. Before we get into the detail, let’s recap the context.
After years of money printing where central banks actively bought government bonds, QE is in reverse. Central banks wish to shrink their balance sheets and are reversing the QE process. They no longer wish to buy bonds. The central banks now wish to exit those positions. The process is called “quantitative tightening” or QT.
But while all central banks are seeing some form of losses, it is the BoE which has the most eye-catching situation due to its outlier status:
(1) The BoE used QE more aggressively than other central banks
When QE was the focus of central banks, the BoE led the charge. Proportionally, the UK central bank bought more government debt than other G3 central banks. This meant that at the peak in 2021 the BoE held more than 37% of all UK government debt. This is a huge figure and nearly twice the level for the US Federal Reserve’s equivalent holding in US treasuries.
It was also more than the figure for the European Central Bank (ECB), despite the ECB having used QE to spend huge sums supporting peripheral bonds – finding many imaginative ways to bring down Italian and peripheral yield levels. But the BOE still ran the printing press hotter.
During QE, the BoE bought around twice as much government debt as the Fed
(Share of government debt held by the central bank, at the peak in 2021)
This is significant. The central banks have to pay cash rates on the reserves created to buy the bonds. The expanded balance sheets of the central banks are financed at overnight rates. The higher rates rise, the more costly this interest expense becomes. The BoE is the most vulnerable to higher losses having used more QE to begin with.
(2) Only the BoE is actively selling
For most central banks, balance sheet reduction is being pursued passively. This is understandable as prices for bonds are today much lower – and yields higher – than when the central banks bought those securities. Inflation and rate hikes have depressed the market. So central banks in general have not wanted to lock in the losses at today’s low bond prices.
For example, at the Fed and the ECB the process of QT simply involves letting the bonds mature. In market parlance, they are “holding until maturity”. Central bank balance sheets shrink steadily as time goes by and the bonds roll off, without any active involvement. This makes sense as there is always the risk that any active sales from the central bank could pressure the market downwards.
But the BoE is doing things differently.
For Threadneedle Street the process very much involves actively selling bonds into the market, in the size and scale that only a central bank can do. This has come at a cost – the extra selling pressure has forced the price of bonds down and yields up – our note published in September estimated the extra supply has pushed gilt yields up by around 0.4%. But the selling has meant a much more rapid shrinkage in the BoE balance sheet.
Back in 2021, the BoE held 37% of all UK government debt. Today, the balance sheet has shrunk such that the figure is down to 28%, i.e. the exit from QE is equivalent to sales of about 9% of outstanding government debt. This is around twice as fast as the pace of the Fed and the ECB.
The BoE is aggressively reversing QE
(Reduction in share of government debt held by central banks, since the peak in 2021)
In our view, there is little benefit to being so hasty. The fast pace of gilt sales pushes down on prices, deepening losses, and worse, crystalises what were paper losses into an immediate drain on the UK taxpayer. It gives no chance for the interest rate environment to improve or the prices for government bonds to recover.
This brings us to the third difference in QE. Uniquely amoung central banks, the BoE has an indemnity that means any loss has to be made good to the government, allowing the Bank to focus on issues other than financial implications for taxpayers.
(3) A legal indemnity: only the BoE requires taxpayer money to make up the losses
Uniquely amoung central banks, the BoE has a legal indemnity against losses from QE. This indemnity was negotiated in 2009 when Alistair Darling was chancellor. Mervyn King, then BoE Governor, was sceptical about QE and insisted the government underwrite any losses. At first these losses never came – instead the zero interest rate environment allowed large profits to be repatriated to the Treasury.
But those days are gone, and the projected losses now dwarf any of the gains made previously. Armed with the indemnity the BoE can take an academic view of these losses, even where its actions (such as actively selling the bonds) are worsening the issue.
It is very different over at the Fed. There the gains and losses are treated as a deferred asset or in the case of losses a liability. Rather than requiring the central bank to be “bailed out” they are simply an accounting entry. This gives the Fed the luxury of time and avoids the rush to sell.
Statements from the Bank suggest it is not concerned about the losses
As far as the Monetary Policy Committee (MPC) goes, QE losses are not their problem. The losses, or ways to avoid those losses, are not something they think about. Perhaps this is due to the indemnity – which ensures the Bank does not suffer the financial consequences.
An example. Recently, BoE governor, Andrew Bailey, told parliament interest rates were at the top of the cycle. If Bailey is right it would be reasonable to expect bond prices to recover in the future. So one might ask, if we are at the peak in interest rates why is the BoE actively selling gilts now (and, indeed, recently stepping up the pace) when bond prices are so low? Deputy governor Ramsden has answered this very clearly: “The MPC does not take into account financial risk or profit when taking monetary policy decisions, including about the gilt portfolio”2.
For the taxpayer, it is a different matter. Bloomberg recently reported calculations by Oxford Economics and Deutsche Bank showing that, relative to simply letting the bonds roll off naturally, active QT will lead to additional losses of around £10 billion a year3. These additional losses will grow if interest rates are cut more than the market currently expects.
Why is the BoE so keen to shrink its balance sheet? The political context matters. Back in 2021, the House of Lords published a report calling QE “a dangerous addiction” 4. At the time, the BoE’s holding of UK government debt was much higher than the Fed or the ECB. This very public critique from the normally restrained upper chamber about the size and scale of the UK’s QE programme made headlines for weeks and jolted the BoE into promising to start reversing the process. The result was the Bank committed to certain targets – including a process of active bond sales – that it now feels it has to stick to.
Arguably, this has become a type of intransigence. Those commitments were made in 2021 when few expected prices for bonds to drop so much, or QT to become so costly. Now, the Bank feels it can’t change course without admitting error and has to demonstrate its willingness to break what the House of Lords called its “additction”, no matter the costs involved.
But beneath the surface, changes could happen
The £170 billion headline cost isn’t yet a certainty. It is an estimate based on certain projections around interest rates and the like. Maybe the Bank could, if not overtly reverse course, tweak the details. That seems to be happening. Recent indications are hinting at certain evolutions in the process of QT.
A hint to this type of evolution was buried in a recent “market notice” report that accompanies the planned sales of gilt holdings5. With new language it left open the possibility of adjusting the maturity composition of the bonds actively being sold. This would allow the Bank to shift the sales towards bonds trading closer to maturity (and closer to par value), and delay the sales of long maturity bonds until interest rates are lower and the low prices have recovered.
Such a step would help. Taxpayers and gilt markets alike would likely breath a huge sigh of relief. But even better would be the Bank adopting the “hold until maturity” stance of the Fed. Who knows, perhaps Ben Bernanke, a former Fed chair currently running a government-appointed review into BoE processes, might quietly mention it.
Alternatively, the scale of the QE losses is likely to draw increasing scrutiny from the Treasury and its representitives. At its meeting on 21 November, the Treasury Select Committee publically asked these questions to Bailey and Ramsden, including where the money to fund these losses would come from. Bailey declined to reply.
In an election year, incentives to bring the BoE in line with other central banks are enormous. Could the Treasury clarify or amend the terms such that the indemnity, while still in place for the type of passive QT pursed by the Fed, should not be used to cover the avoidable losses from active QT? Such an announcement would immediately result in a sharp rally in government bonds as the biggest seller of gilts would disappear from the market. This would lower gilt yields and correspondingly lower the government’s projected interest costs over the next many years. In otherwords, fiscal headroom for any pre-election tax cuts.
For now, such ideas remain speculative. Either way, in our view the gloom hanging over the gilt market is a case of it being “darkest before dawn”. As we have written before, we think that in the same way the BoE gold sales of the early 2000s marked the low of the gold cycle, the current intensity of the Bank’s bond sales will mark the nadir of the gilt cycle.
Even without the tweaks and evolutions to QT policy, the worst for the gilt market is behind us. Inflation is already coming down. Next year will see even more convincing falls in CPI and much more talk of interest rate cuts. 2024 should be the year of fixed income.