Reactions to the Bank of England’s gilt market operation

Authors: Professor David Aikman and Dr Richard Barwell

Published: 11 October 2022

The Bank of England’s “temporary and targeted financial stability operation” proved highly effective at stabilising the gilt market. The Bank managed to break the feedback loop that had emerged, whereby margin calls triggered sales of gilts which in turn further depressed prices, and its intervention catalysed a rapid recovery in gilt prices.  In so doing, the Bank averted what might otherwise have quickly morphed into a full-blown crisis for the pension industry, the financial system and ultimately the economy.  So the intervention was undoubtedly a success from a crisis management perspective.

However, we should not evaluate the ultimate success or failure of the intervention purely on whether it was able to arrest and reverse a destabilising decline in gilt prices on day one. After all, it is not so surprising that an unexpected announcement that the central bank will print money and buy bonds would support bond prices. The Bank will need to reflect on whether the design and execution of the intervention are well aligned with the objectives of the plan and relatedly whether the intervention was well understood by market participants – or put another way, whether bond prices recovered for the right reasons. It will also need to successfully execute the exit strategy. These issues are the focus of this blog.

Photo by Ryan King from iStock

The drawback with “buyer of last resort” QE

This is by no means the first time central banks have purchased assets for financial stability purposes.  The massive purchases in Spring 2020, when central bankers acted as a “buyer of last resort”, are a recent case in point. There is no doubt about the restorative impact these purchases had, in what were highly dysfunctional markets.  Nonetheless, with the benefit of hindsight, central banks might have tweaked the design of those programmes (Barwell (2022)). While they had both monetary policy and financial stability aims, financial stability policymakers did not appear to play a major role in their calibration.  Nor was there any clarity on how much of the bond buying was driven by each objective. After the fact, a common reinvestment policy – the one dictated by monetary policy considerations – was applied to all the purchases, so effectively the purchase programmes morphed into entirely monetary policy operations.

This is not ideal because it feeds the perception that all asset purchases are ultimately the same thing – Quantitative Easing (QE).  This complicates the communication strategy the next time a central bank wants to engage in an asset purchase programme for financial stability purposes. This is particularly problematic in a situation where the direction of travel of monetary policy is towards a tighter stance and hence where a financial stability intervention of this kind might be working at crossed purposes with monetary policy. Of course, that is exactly the situation in which the Bank found itself in late September.

A financial stability intervention: “temporary and targeted”

One criterion on which the Bank’s September intervention should be evaluated is whether its design and execution is clearly delineated from a classic monetary policy operation.  There are three reasons to be positive here:

First, the Bank made it clear in its communication that this was a financial stability intervention. The press release explains that the intervention is designed to “tackle a specific problem in the long-dated government bond market” with the aim to “restore orderly market conditions”. Consistent with the idea of targeting purchases on the source of the problem – as opposed to an attempt to ease broad financial conditions – the Bank would later announce that the scope of the operations was to be widened to include purchases of index-linked gilts, given the evolving price action in that segment of the market. Of course, we understandably do not have a quantitative definition of what constitutes an orderly market so it is not immediately obvious how to assess whether the Bank has achieved its stated objective or not.

Second, guidance around the pace of purchases and their eventual unwind conforms with a financial stability intervention. The pledge to buy on “whatever scale is necessary to effect this outcome” was undoubtedly powerful, as evidenced by the reaction of gilt prices.  Likewise, the pledge to reverse the purchases “in a smooth and orderly fashion once risks to market functioning are judged to have subsided” is an essential feature which prevents the purchases morphing into a monetary policy programme after the fact.  Once again, there is no definition of what constitutes a smooth and orderly unwind – and hence how long the bonds will stay on the balance sheet – or how the Bank will judge when the risks have subsided. As we shall go on to discuss, if the latent fragility that triggered the sell-off remains, it is not clear that the risks to market functioning will have subsided.

Third, ownership of the programme resides with the Financial Policy Committee rather than the Monetary Policy Committee, although the Governors of the Bank remain the principal policy actor where the balance sheet is concerned. To be precise, we read in the Bank’s press release that “on 28 September, the Bank of England’s Financial Policy Committee noted the risks to UK financial stability from dysfunction in the gilt market. It recommended that action be taken, and welcomed the Bank’s plans” whereas “the Monetary Policy Committee has been informed of these temporary and targeted financial stability operations. This is in line with the Concordat governing the MPC’s engagement with the Bank’s Executive regarding balance sheet operations”.

An unsterilised intervention and a temporary pause on asset sales

Despite this, questions remain.  If this was a pure financial stability operation, as the Bank argues, why was the decision not sterilised?  That is, why did the Bank allow it to expand the stock of central bank reserves?  And given it did expand the stock of reserves, why was the MPC only “informed” and not consulted?

Combined with the decision to put the scheduled programme of bond sales – or active Quantitative Tightening (QT) programme as it is known – on hold until the end of October, this has placed the stock of reserves on a different trajectory to the one that was envisaged only a week earlier at the time of the September MPC meeting – at least for a time.  The perception that the stance of monetary policy had swung from QT to QE may therefore have taken hold in some quarters.  After all, if the Bank is printing money to buy bonds with the intention of driving down gilt yields, then that sounds very much like what it has been doing periodically over the last decade to achieve the inflation target.

In the Bank’s defence, it would not have made much sense to simultaneously commit to buy bonds on whatever scale is necessary to stabilise the gilt market whilst at the same time selling bonds back into the market.  Putting active QT on pause for a few weeks makes sense. The Bank has also communicated that the intervention will not have a lasting impact on the stock of asset purchases or reserves. The press release states that “the MPC’s annual target of an £80bn stock reduction is unaffected and unchanged”. Chief Economist, Huw Pill, argued that “these operations do not create central bank money on a lasting basis” (Pill (2022)). Whether “on a lasting basis” ultimately turns out to mean weeks, months, quarters or even years will play a key role in determining the ex post evaluation of this scheme and hence the ex ante response to future announcements of this kind.

The acid test of the programme is therefore whether it affects the stock of central bank reserves on a lasting basis. If the Bank wants to firmly establish the concept of temporary asset purchases as a tool of financial stability in the minds of investors, then it needs to deliver on the pledge to unwind the purchases. Greater clarity over what the key phrases identified above mean – “in a smooth and orderly fashion once risks to market functioning are judged to have subsided” and “on a lasting basis” – will be important in improving understanding of the programme within markets and anchoring market expectations.  But actions speak louder than words. If the unwind is glacial, then it will become harder to convince some sceptical investors that the Bank did not indulge in a top-up of the (monetary policy) QE programme.

Is this a conventional QE programme in all but name?

One practical question is whether the Bank’s temporary and targeted intervention should be expected to have a similar impact to those earlier, conventional QE programmes. There are two schools of thought on the mechanism through which conventional asset purchases influence long-term bond yields and ultimately inflation: either they signal something about the reaction function of the committee that sets monetary policy or they persistently disturb the balance between demand and supply in the bond market. A temporary purchase programme that was not instigated by the MPC and is soon unwound does not neatly fit either story.  The MPC cannot be sending a signal if it didn’t design the programme and – so long as the programme proves temporary – then there is no lasting disturbance to the balance between demand and supply.

What really matters is that this financial stability intervention does not interfere with the conduct of monetary policy (Tucker (2009)).  Hence, the MPC must be free to adjust the stance of policy as it sees fit in light of the impact of the intervention on gilt prices, broad financial conditions and ultimately the outlook for inflation. The Bank’s statement is clear on this point: “The MPC will not hesitate to change interest rates by as much as needed to return inflation to the 2% target sustainably in the medium term, in line with its remit”.  It will be important for the MPC to demonstrate that it will do what is necessary to achieve the inflation target. One key aspect of ‘walking the talk’ on what the ECB calls the “separation principle” will be the resumption of active QT at the end of the month.

Why sterilisation matters and classifying the scheme

The Bank could have sterilised this programme by issuing short-term liabilities to drain the reserves that were created back out of the system.  To be clear: in terms of the economic impact of the programme this would have been a largely cosmetic change. From the perspective of the balance sheets of the consolidated state and the private sector, the intervention is an asset swap, with the central bank exchanging a very short-term liability (reserves) for a very long-term liability (a long-dated gilt). The economic impact is little changed if the central bank then engages in an additional tidying up exercise, with the central bank issuing a new very short-term liability (Bank of England paper) or selling bonds from the QE portfolio with a very short residual maturity, in order to drain another very short-term liability (Bank of England reserves), particularly in an environment where the central bank has previously flooded the system with the latter through a decade of QE.

However, optics matter.  Insisting on sterilisation would have made it harder for the “monetary U-turn: QT to QE” meme to take hold, and that is very important.  It would also have been a little easier to classify the scheme as a classic market maker of last resort operation had the purchases been sterilised.  Indeed, it is interesting that the Bank did not choose to use that terminology to describe the intervention.

Barwell (2022) offers a theoretical way to distinguish between different types of financial stability intervention designed to put a floor under asset prices. In principle, the market maker of last resort (MMLR) leans against excessive liquidity risk premia in asset prices whilst the risk taker of last resort (RTLR) leans against a broader set of factors that might weigh on asset prices in a crisis. When prices are falling fast there may be no time to conduct a clean and coherent identification of the contribution of liquidity risk premia to the sell-off. Nonetheless, it is a useful exercise to ask oneself the question: is there any desire to lean against an excessive and unwarranted inflation risk premia or real term premia should either become embedded in gilt prices in this scenario?  The Bank’s public line is clear: Deputy Governor Cunliffe’s letter to the Treasury Committee insists that these operations are not “designed to cap or control long-term interest rates. Their intention is instead to ensure that those yields are not distorted by severe liquidity strains in financial markets”.

Nonetheless, the decision to “buy time” and cap the rise in government bond yields following a fiscal announcement has caused concern in some quarters.  Jason Furman, a highly respected American economist, argues that “this is how fiscal dominance starts”. Indeed, the Bank has chosen to conduct a form of “yield curve control in the tails”, if only temporarily. However, one could reasonably ask what a responsible central bank should otherwise have done?  The Bank cannot threaten finance ministers that it will allow a financial crisis through inaction if they do not change fiscal course.  The Bank had no choice but to act. Some form of yield curve control in the tails is always implicitly there:  if markets are disorderly then a central bank will always come under pressure to act.

Will the Bank be able to step back?

The Bank has stated that the purchases will be “strictly time limited” with auctions taking place until 14 October.  At face value, this statement was reassuring and is consistent with the nature of the asset purchase programme: financial stability interventions should be “temporary and targeted”.  The subsequent announcement of larger gilt auctions and a Temporary Expanded Collateral Repo Facility (TECRF), which should allow banks to ease the liquidity squeeze on LDI funds, were no doubt intended to reassure the market that the exit strategy was achievable.

Unfortunately, the Bank is not necessarily in a position to make unconditional statements about when gilt purchases will stop and when gilt sales will resume because that depends on market conditions.  As with all such statements, there is a question about the credibility of the commitment: whether the Bank can step back, once it has stepped in to prop up the gilt market. Larry Summers for one, is not convinced, arguing that “I think lenders of last resort make a mistake — well, financiers of last resort make a mistake — when they declare how time-limited their operation is going to be. And I would not be surprised if the Bank of England finds it necessary to adjust its position.”

Indeed, the Bank’s statements this week that it “plans” to end these operations on Friday 14 opens the door to the possibility that these operations will remain in place beyond that original end date.  Deputy Governor Dave Ramsden described the intervention as “an operation designed to buy time” (Ramsden (2022)). The Bank just may need more time than it originally hoped.  The same logic may unfortunately apply to the resumption of active QT at the end of the month. After all, if the Bank reaches the conclusion that the gilt market will soon sink back into dysfunction once the purchase programme expires, then it is hard to believe it will be able to withdraw support from the market for any length of time.

No doubt the Bank and others are working very hard behind the scenes to to ensure that there has been a sufficient injection of capital or reduction of leverage by the 14 October that would allow the Bank to step back. Indeed, Deputy Governor Cunliffe confirms that “the Bank, TPR and the FCA are closely monitoring the progress of LDI funds as they take action to put their positions on a sustainable footing for whatever level of asset prices prevails at the end of the operation and to ensure LDI funds are better prepared for future stresses given the current volatility in the market”. The latest statement from the Bank confirms that “LDI funds have made substantial progress ..  over the past week” but it also sends a very clear warning: “the beginning of this week has seen a further significant repricing of UK government debt, particularly index-linked gilts. Dysfunction in this market, and the prospect of self-reinforcing ‘fire sale’ dynamics pose a material risk to UK financial stability.”

The Bank may also have concluded that a deadline on the purchases was a necessary ingredient to focus minds on these efforts to reduce leverage and raise capital. Working against this argument somewhat is the fact that the initial decline in gilt yields that the Bank engineered might have, other things equal, reduced pressure on the LDI managers to act. The belief that the Bank can be relied upon to keep buying bonds to cap yields would further reduce the incentive to act.  The Bank has tried to tie its hands to the mast and pledge that the support is time-limited and has arguably used its own reputational capital as a commitment device. It would be unfortunate if it had to perform a U-turn.

The broader problem for the Bank, and indeed the wider central banking community, is that here we are again: a fragility in the non-bank financial sector has forced an emergency central bank intervention to put a floor under asset prices; this in turn has further entrenched beliefs in the central bank put. The case for more intense surveillance of the non-bank financial sector and policy interventions to address its deep underlying vulnerabilities is as strong as ever.