The return to normality

Quantitative easing (QE) is in retreat – marking the end of a near-15-year financial experiment. What began as an unconventional monetary policy tool rapidly became a policy linchpin that saw central banks amass huge balance sheets, which mushroomed again as Covid-19 hit.

But now the Bank of England (BoE), Federal Reserve and European Central Bank (ECB) are going into reverse, ushering in a new experiment in the process: they are embarking on programmes of quantitative tightening (QT) that will see balance sheets start to ‘normalise’. QE saw central banks buy longer-maturity bonds using digitally created reserves, thereby boosting liquidity and stimulating economies. Figures from Goldman Sachs Wealth Management estimate that markets will need to absorb $1tn in net bond supply as part of the reversal of this process. Just as QE lowered bond yields, economists expect potential bond buyers to demand higher yields in the face of this growing supply. 

This is not the only concern. In the US, an earlier attempt at QT was abandoned in 2019 as overnight borrowing costs soared, despite Fed chair Janet Yellen’s insistence that QT would be as boring as “watching paint dry”. There remains a risk that QT could have an outsized effect on liquidity conditions and even pose a threat to financial stability.

But the impact could also prove more benign, especially given central bankers’ determination to unwind QE both slowly and predictably. By tightening financial conditions, QT could also mean that central bankers do not need to raise interest rates by as much as expected. The policy then could either cause fireworks or simply fizzle out this year.

Why unwind QE?

There are several reasons why central banks are keen to roll back QE in the coming months. Firstly, by reducing liquidity and contracting balance sheets, the shift will have a monetary tightening effect. This should complement the inflation-busting impact of higher policy rates – welcome news for central banks wrestling with stubbornly high price levels. 

A June study by Fed economists estimated that reducing the Fed’s balance sheet by $2.5tn over the next few years (for context, $1.6tn is expected over the course of 2022 and 2023) would have the same impact as raising the policy rate by 50 basis points (bps). OECD economists came to similar conclusions, tentatively estimating that QT will increase long-term interest rates by between 50 and 100bps in the US, and between 30 and 80bps in the UK. 

This is not immaterial, but it seems likely that the impact of QT will be overshadowed by the sharp increases in policy rates seen in 2022 and in the year ahead (see chart). Nevertheless, the additional impact – if QT proves to be a way of tightening policy less severely than by raising base rates – could allow central banks to forego a rate hike or even begin lowering rates earlier if necessary, a boon for economies teetering on the brink of recession. 

These rising policy rates have created another rationale for the big unwind. Generally, under QE, bank reserves (floating rate debt), are used to purchase gilts (long-term fixed-rate debt) which increases the sensitivity of interest payments to changes in the interest rate. As policy rates rise, they start to exceed the interest earned on central bank bond holdings. Ex-BoE monetary policy committee (MPC) member Michael Saunders argued in December that a strong rationale for QT is to ensure that the maturity profile of public debt and its exposure to short-term interest rates “gets back to where we would want it to be”. 

Central banks may find themselves crystallising losses as gilts mature or are sold off. In the UK, the BoE purchased many of its gilts above par but is selling them below par: yields have risen significantly since it purchased the gilts, meaning prices have moved inversely. The Office for Budget Responsibility (OBR) estimates that the Treasury may need to transfer £133bn to cover these losses over the next five years.

These losses are unlikely to create financial stability concerns for the Fed, BoE and ECB, but do put central banks in a difficult position. Analysts at fund manager Janus Henderson note that the ‘optics’ of Treasury departments sending money to central banks as they sell bonds at a loss could “incite some public backlash” as well as raising questions about central bank reputation and independence. 

QT in the UK 

By international standards, the BoE has moved proactively to unwind its easing programme. In February 2022, so-called ‘passive QT’ began, when the Bank stopped reinvesting the proceeds from maturing bonds, allowing it to ‘run off’ the balance sheet instead. In November, the UK became the first country to start ‘active QT’ when the central bank began selling bonds back to financial markets.

The Bank stresses that QT will be orchestrated “in a gradual and predictable manner”, and the scale of the unwinding has so far been relatively small. The outstanding stock of bonds fell by £44bn over 2022 as a whole, and just £6bn of this was the result of ‘active’ gilt sales in the fourth quarter. Over the year ahead, the Bank aims to reduce its stock of gilts by £80bn through a combination of ‘passive’ maturing bonds and ‘active sales’ of around £10bn a quarter. 

The UK’s ‘active’ approach to QT is not the only factor making it an outlier. The chaotic market reaction triggered by September’s mini-Budget saw 10-year yields soar to 4.5 per cent, and the BoE was forced to delay sales of long-end bonds and revert to something akin to QE in order to stabilise markets. 

Ten-year yields have since retreated to around 3.3 per cent, but Saxo Bank analysts have warned that “further pressure on gilts cannot be [ruled out]”.

In December, the BoE announced that it would resume sales of long-maturity bonds this month as part of its QT operations. Prices fell and 10-year yields climbed by 17bps in response, with investors speculating that increased supply could drive prices down further. Edward Hutchings, head of rates at Aviva Investors, told the IC that “although in theory it should make little difference”, active QT is likely to have a greater impact than passive tightening as the market is forced to “make room for these extra bonds by pricing yields at higher levels”. 

QT at other central banks 

Other major central banks are sticking to the passive approach to tightening.  

In the US, monetary stimulus programmes saw the Fed purchase both US Treasury bonds (which made up around two-thirds of asset purchases) and mortgage-backed securities by injecting reserves into the banking system. The Fed ended QE in March 2022, and started passively reversing the process in June by not reinvesting up to $30bn in maturing Treasuries and $17.5bn of maturing MBS every month. Those caps rose to $60bn and $35bn, respectively, in September.

Unlike the UK, the US has dabbled in QT before – with limited success. Between October 2017 and September 2019, the Fed’s balance sheet unwinding came to an abrupt end after bank reserves dropped below the minimum level needed to ensure the smooth functioning of short-term lending markets. Fears of a repeat performance could prevent the Fed from moving further this year. Analysts at fund manager T Rowe Price argue that the effective level of minimum reserves is now even higher than during the previous tightening attempt.

In the euro area, the shape of QT looks different again. The ECB plans to start passive tightening at a “measured and predictable pace” from March 2023, when it will cease to reinvest maturing securities up to a total of €15bn per month. 

But matters are complicated by the fact that not all European bonds are equally risky. After the ECB announced its plan to commence QT in December, eurozone bond yields rose sharply. The spread between German and Italian 10-year yields rose to 220bps, the highest in over a month, reflecting higher premiums on ‘riskier’ Italian debt. 

Vítor Constâncio, a former vice president of the ECB, has voiced concerns about the possibility of “financial fragmentation”, given the heterogeneous nature of euro area economies. If realised, this would risk a repeat of the 2010 euro area sovereign bond crisis, and ultimately impair the transmission of monetary policy. Aviva Investors’ Hutchings expects “peripheral bonds to struggle from here and European bonds in general to be somewhat less supported, and particularly so in light of the glut of bond supply to hit financial markets next year”. 

There is also the issue of liquidity. Tamara Basic Vasiljev, senior economist at Oxford Economics, told the IC that shortages of some securities (German bonds in particular) and the ECB’s regulatory framework have made some market segments extremely illiquid, complicating matters further.

Not simply a reversal

Though central banks have clearly signposted the pace and magnitude of QT, there is much less clarity about the impact it will have. 

Evaluating the impact of monetary policy is difficult at the best of times, and central banks are currently raising interest rates in the face of uncertain transmission mechanisms and time lags. With QT, the challenge is even greater. Stephen D Williamson, economist at the Federal Reserve Bank of St Louis, notes that when it comes to the impact of unconventional monetary policy, “economic theory is lacking and there is a small amount of data available for empirical evaluation”. 

But what little we do know about the process indicates that it will not simply be a case of ‘QE with the signs reversed’. Context seems to be key. The easing policy was primarily used at times of crisis, meaning it happened suddenly, taking markets by (pleasant) surprise. When it comes to withdrawing liquidity, central banks are at pains to move slowly and predictably. Ex-MPC member Michael Saunders stressed in December that the Bank is “not seeking to disrupt markets, not seeking to push gilt yields higher”. He added that “Bank Rate is the main monetary policy tool, and QT is likely to sort of chug along in the background – hopefully, as a low-drama exercise”. 

The clear signposting about the scope and pace of the process means that much of today’s tightening has probably already been ‘priced in’. ING senior rates strategist Antoine Bouvet notes that  “moves have been well-telegraphed months in advance, and so we’ve already seen part of the increase in yields that this should trigger”. TS Lombard head of macro research Freya Beamish adds that while “QT is a powerful drag on asset prices…the plans are known”. 

Nor will it represent a full ‘reversal’ of the policies central banks have undertaken over the past 15 years. Vasiljev argues that “it is very unlikely that central banks will ever dare” to push balance sheets back to pre-pandemic levels” (let alone beyond) given the potential impact on asset markets. Her models suggest that every trillion dollars of QE pushed the S&P 500 up by up to 20 per cent. By this estimate, “most of the gains since pandemic recovery were QE-generated, and this will most likely make full QT impossible to achieve”. 

At a global level, OECD research suggests that the tightening will proceed at such a slow pace that by the end of 2024 central bank holdings will still be at a higher level than before the pandemic, both in nominal terms and as a share of GDP. 

Macroeconomic backdrop matters 

If QT is designed to go on in the background, we may well find that its impacts will be overshadowed by whatever comes to the economic ‘foreground’ next year. 

The US’s limited past experience suggests that macroeconomic performance is key. Morgan Stanley analysts note that in 2018-19, bond yields fell – despite QT. This is not the reaction we might expect: quantitative tightening increases bond supply, putting upward pressure on yields. They attribute the topsy-turvy impact to the countervailing forces – of moderating growth and inflation – that were then in effect. 

In the euro area, Morgan Stanley analysts anticipate that a combination of lower than expected interest rates and declining inflation will “more than offset the negative impact of ECB QT”. But much remains unclear: these forecasts come with the strong warning that “with little historical precedent for QT, we are sceptical that anyone can make predictions with much statistical precision”. 

Any negative impacts on liquidity could also be offset by global forces. TJ Scavone, investment director at investment consultant Cambridge Associates, warns that if liquidity conditions deteriorate enough to risk financial stability “central banks may be forced to prioritise relieving short-term pressures over fighting inflation”. These fears are not overblown: he adds that this is, to some extent, what happened with the Bank of England last year.  

But Michael Howell, managing director at Crossborder Capital, offers a more optimistic vision for global liquidity, expecting it to pick up as China stimulates its lockdown-hit economy. He notes that it is also possible that a weaker dollar could increase the take-up of cross-border loans, while weaker oil prices should reduce the amount of credit required to finance transactions. 

QT and policy rates 

It is likely that the relationship between QT and policy rates will be an increasingly important theme this year. Though changes to policy rates and unconventional stimulus policies both impact interest rates, they operate in different ways: policy rates target shorter-term market rates while central banks tend to use QE to target longer-term rates.

Central banks have made it clear that interest rates will remain the primary policy tool of monetary policy, and it is not hard to see why: analysts at Janus Henderson note that “we know from history how higher interest rates affect economies, but we don’t for balance sheet reduction”.

It seems unlikely that QT will continue once central banks move to cut policy rates. Janus Henderson analysts argue that reducing a balance sheet at the same time as lowering rates would be “akin to putting a foot on the accelerator and brake at the same time”. With rate cuts expected towards the end of the year (see chart), QT could prove distinctly more short-lived than its expansionary counterpart. Seth Carpenter, chief global economist at Morgan Stanley, expects the first US rate cut to come in December 2023, and for QT to end next year.

Yet Carpenter says that there is a possibility the end could come even earlier if the economy goes into recession, triggering the kind of aggressive rate cuts that would not be compatible with the continued unwinding of monetary stimulus. He also highlights the risk that the process is forced to end prematurely if markets become dysfunctional. ING rate strategist Antoine Bouvet argues that the potential financial stability risks posed by QT will leave central banks with less incentive to carry on with it once their policy focus shifts to easing. There remains a chance that this tightening could block up financial plumbing.

Out with a whimper or a bang?

Cambridge Associate’s Scavone expects policymakers “to respond to severe bouts of financial market stress”, which could see assets rebound sharply, even if their threshold for what constitutes stress proves higher than in the past. Scavone advises that testing portfolio liquidity will help ensure that portfolios are well-diversified, and that they are able to take advantage of any opportunities that arise during periods of market stress. 

Yet Scavone cautions against “any significant changes to portfolio allocations in response to QT” – or indeed any other tail risks. After all, if it progresses as planned, the whole thing should indeed be as tedious as watching paint dry – with changes to policy rates taking centre stage. 

It is also worth noting that this period of tightening could, paradoxically, increase the chances of a return to stimulus measures in the years ahead. Economists at the Richmond Fed argue that if central bank balance sheets continued to grow, policymakers could have, in theory, run out of bonds and other acceptable assets to purchase to conduct future rounds of QE. The Bank of England certainly hasn’t ruled it out, with chief economist Huw Pill suggesting a hypothetical return to bond purchases in a few years' time once policy eases. Engaging in QT frees up capacity for future stimulus, allowing central banks to ‘reload’ and build firepower for the next crisis. Watch this space.