Another decade of ultra-loose monetary policy - Capital Economics
UK Economics

Another decade of ultra-loose monetary policy

UK Economics Focus
Written by Paul Dales
Cancel X

The swift and significant response of the Bank of England to the coronavirus crisis has prevented a financial crisis, but we think the Bank will need to do much more than the markets currently expect to get the economy back on track. By this time next year, we suspect that the Bank will have announced an extra £350bn of quantitative easing, which would take the stock of QE to almost £1,000bn. We think it will be five years before the Bank raises interest rates above 0.10% and at least ten years before it even thinks about unwinding QE.

  • The swift and significant response of the Bank of England to the coronavirus crisis has prevented a financial crisis, but we think the Bank will need to do much more than the markets currently expect to get the economy back on track. By this time next year, we suspect that the Bank will have announced an extra £350bn of quantitative easing (QE), which would take the stock of QE to almost £1,000bn. We think it will be five years before the Bank raises interest rates above 0.10% and at least ten years before it even thinks about unwinding QE.
  • Most analysts and market participants expect the Bank of England to build on the rate cuts from 0.75% to 0.10% and extra £200bn of QE announced in March by expanding QE by £100bn at its next policy meeting on 18th June. They also believe that interest rates will be 0.10% or below for the next three years.
  • Our forecasts that by the end of 2022 GDP will still be about 5% lower than it would have been if the coronavirus never existed, the unemployment rate will be closer to 5.5% than the pre-crisis rate of 3.8% and CPI inflation will still be well below the 2% target suggests the Bank will need to do more.
  • We estimate that the Bank will need to loosen policy by the equivalent of a 250 basis point (bps) cut in interest rates. Coming on top of the loosening equivalent to a 200bps rate cut put in place in March, that would equate to a total stimulus worth 450bps and would compare to the 750bps of action during the GFC.
  • We’re assuming that all of this extra loosening will be achieved by the Bank expanding QE by an extra £350bn in total, perhaps in four separate steps spread over the next year. About £330bn of that may be extra purchases of gilts and about £20bn additional purchases of investment grade corporate bonds. This would take the outstanding stock of QE to £995bn – just shy of £1trn and equivalent to 45% of GDP.
  • But we wouldn’t rule out the Bank also resorting to other tools at some point. Negative interest rates would be a big symbolic step, but wouldn’t be a magic bullet as they can probably provide only an extra 100bps of policy space at most. And given the concerns that they could trim bank profits and limit the supply of credit, the Bank would use them with caution. Rather than cut rates quickly to -0.5% or -1.0%, it would probably move in baby steps over a long period of time.
  • Equally, the Bank may consider buying riskier assets, such as non-investment grade corporate bonds, or adopting yield curve control, which involves setting a target for gilt yields. But the Bank is reluctant to take on the default risk associated with the former and the latter may only become appealing if concerns over the sustainability of public debt start to push up gilt yields, which we are not expecting.
  • It makes sense that the Bank would become more experimental the longer the economy needs support. And the risk is that even our relatively downbeat forecasts for the economy prove to be too optimistic and that the Bank ends up loosening policy by even more than we are expecting.
  • Perhaps the bigger issue is that monetary policy is just not as effective at boosting demand and inflation as it once was. That suggests monetary policy will have to remain loose for many years and that fiscal policy will need to do much of the heavy lifting. We think interest rates won’t rise above 0.10% for five years and that it will be at least ten years before the Bank of England is in a position to even think about unwinding QE.

Another decade of ultra-loose monetary policy

The Bank of England should be congratulated for its speedy policy moves in March that appear to have successfully prevented the economic fallout from the coronavirus crisis from spiralling into a much more damaging financial crisis. But we think the Bank will need to do more over the next year or so. This UK Economics Focus explains why, what form any further stimulus may take, when it may happen and how effective it might be.

What has the Bank done so far?

Table 1 summarises the actions taken by the Bank at its two emergency meetings in March. In short, interest rates have been cut from 0.75% to a record low of 0.10% and QE was expanded by £200bn (£190bn of extra gilt purchases and £10bn of extra corporate bond purchases). Amongst other things, the Bank also established the Term Funding Scheme with incentives for Small and Medium-Sized Enterprises (TFSME) to provide cheap funding for banks and the Covid Corporate Financing Facility (CCFF) to provide funding directly to businesses.

Table 1: Bank of England Recent Policy Action

Measure

Size

Aim

1. Bank rate cut

From 0.75% to

0.10%

Help economy to rebound

2. QE

£200bn (£190bn

gilts, £10bn

corporate bonds)

Ease market stress. Help economy to rebound

3. Decrease in cost and rise in volume of loans in other currencies

UK banks can

borrow US dollars

at a rate only just above zero

Reduce market stress

4. Term Funding Scheme Small and Medium-Sized Enterprises

Unlock at least £200bn of finance

Incentivise bank lending to SMEs

5. Reduction in countercyclical capital buffer

From 1% to 0%. Unlocks up to

£190bn of capital

Free up bank lending

6. Covid Corporate Financing Facility

Unlimited

Loans to businesses

Sources: Bank of England, Capital Economics

These moves were partly designed to boost liquidity and fulfil the Bank’s role as lender of last resort. That was particularly the case with the £200bn of extra asset purchases announced in mid-March when the stresses in the financial markets send gilt yields higher. But the moves were also designed to support the economy and help the Bank hit the inflation target in two years’ time. (See here.)

The low starting point for interest rates means that the 65bps reduction in rates is far smaller than the 525bps of cuts the Bank implemented during the Global Financial Crisis (GFC). The US Fed has only been able to cut rates by 150bps this time compared to 500bps in the GFC. The ECB has only cut its deposit rate by 10bps. (See Chart 1.)

Chart 1: Policy Rates (%)

Source: Refinitiv

But both the size and speed of the Bank of England’s expansion of QE has been impressive. £200bn is by far the largest quantity of QE the Bank has announced in one go. (See Table A1 in the Appendix.) What’s more, the Bank has been buying gilts at a pace of around £60bn per month compared to its previous top speed of £30bn a month. This has resulted in the stock of QE jumping from 20% of GDP to 25% in the last two months alone. That’s similar to the scale and speed of expansion in QE elsewhere. (See Chart 2.)

Chart 2: Stock of QE Purchases (As a % of GDP)

Source: Refinitiv

We can attempt to illustrate how much policy has been loosened in total by using the Bank’s previous estimates of how much QE is equivalent to a 25bps interest rate cut. The declining potency of QE means that the Bank reduced its estimate of effectiveness of QE from £22bn of asset purchases in 2009 being roughly equivalent to a 25bps cut to about £36.5bn in 2011/12. If we use the 2011/12 estimate (which possibly overstates the effectiveness now), then the £200bn of QE is equivalent to a 135bps rate cut. Add in the actual 65bps of rate cuts and so far the Bank has loosened policy by the equivalent of a 200bps rate cut. That compares to the total loosening of around 750bps during the GFC.

How much looser?

There is no way of knowing exactly how much further policy needs to be loosened, but we think the equivalent of another 250bps rate cut is in the right ballpark.

Admittedly, some analysts have suggested that to prevent gilt yields from spiking like they did in the middle of March, the Bank will need to buy as many gilts as the government issues. Our forecasts that the government will have to issue a total of £965bn of gilts over the next five financial years could therefore suggest that the Bank may have to increase QE by an extra £765bn (£965bn less the £200bn of QE already announced). (See Chart 3.) That would be equivalent to a rate cut of around 525bps.

Chart 3: CGNCR & Net Gilt Issuance (£bn)

Sources: HMT, Capital Economics

But with the stress in the financial markets having eased since the middle of March, we think that the prospect of low interest rates and low inflation will keep gilt yields low regardless of how many assets the Bank purchases. As such, we believe the Bank has largely fulfilled its role as lender of the last resort (although its liquidity and funding schemes will continue to grow). After all, this is what has happened during the previous bouts of QE. Initially the Bank buys as many gilts as being issued, just as it has been doing in recent weeks. But after a few months, the Bank’s asset purchases fall relative to the additional gilt issuance. (See Chart 4.)

Chart 4: Gilt Issuance & BoE Purchases (£bn Per Mth)

Sources: DMO, Bank of England

Instead, the bulk of any further action will probably be driven by a motivation to boost demand and meet the 2% inflation target. Of course, there’s not a lot the Bank could have done to support demand during the coronavirus lockdown. That’s why the onus was on fiscal policy to try and keep businesses afloat and employees in work.

But with the lockdown now being eased, the Bank can try and boost demand. At face value, our forecast that GDP will take two years to reverse the 25% fall that we think happened in the first two quarters of this year suggests that by the end of 2022 GDP will be around 5% lower than if coronavirus never happened. (See Chart 5.)

Chart 5: Real GDP (Q4 2019 = 100)

Sources: Refinitiv, Capital Economics

If everything else was unchanged, that would equate to an output gap (the gap between actual GDP and potential GDP) of around 5% of GDP. Monetary policy can and should be used to close that gap. A Taylor Rule shows how the Bank has in the past set interest rates in response to deviations in inflation from the 2% target and the size of the output gap. It shows that an output gap of 5% corresponds to a further easing in policy equivalent to about a 350bps reduction in interest rates. (See Chart 6.)

Chart 6: Actual & Implied Bank Rate (%)

Sources: Refinitiv, Capital Economics

However, the required loosening is unlikely to be quite that large as potential GDP (or supply), has probably fallen. At least some of the reduction in overseas travel and fall in visits to pubs, cinemas and restaurants triggered by the coronavirus lockdown will be permanent, rendering some businesses and services obsolete. If we were to assume that around a third of the gap between the black and blue lines in Chart 5 is due to a reduction in supply, then the resulting output gap of between 3.0% and 3.5% of GDP would suggest that the Bank may need to loosen policy by the equivalent of a 250bps rate cut.

Coming at it from another direction provides a similar result. Our forecast that CPI inflation will be closer to 1% by the end of 2022 than to 2% is also a result of demand remaining persistently weak relative to supply. (See Chart 7.)

Chart 7: CPI Inflation (%)

Sources: Refinitiv, Bank of England, Capital Economics

Based on the Bank’s econometric model that suggests every 50bps cut in interest rates raises inflation by 25bps in two years’ time, our inflation forecast suggests that policy may need to be loosened by the equivalent of a 200bps rate cut. Given that policy may now be less effective than the Bank’s historical model suggests, we think it is reasonable to assume that policy may need to be loosened by the equivalent of a 250bps rate cut.

Coming on top of the loosening worth around 200bps already put in place, then during the coronavirus crisis in total the Bank may loosen policy by the equivalent of a 450bps rate cut. That would dwarf the loosening seen during the 2011/12 euro-zone debt crisis and after the 2016 EU Referendum. But it would be smaller than the loosening of 750bps during the GFC. (See Chart 8.)

Chart 8: Recent Monetary Policy Easing Cycles (Equivalent Reduction in Bank Rate, ppts)

Source: Capital Economics

The Bank could achieve such a further loosening in policy by using some or all of the following tools:

  • Forward guidance.
  • Negative interest rates.
  • Quantitative easing.
  • Credit easing.
  • Yield curve control.
  • Helicopter money.

Forward guidance

This is unlikely to account for any of the 250bps of extra policy loosening. After all, since short-term interest rates are already so low, there is little that the Bank can do to lower them further. Admittedly, it could try to get a bigger bang for its buck by returning to “state-contingent” forward guidance, which involves future interest rate changes being linked to the evolution of a specific economic variable. But the Bank got its fingers burnt by this in 2013 and 2014.

Shortly after Mark Carney took over as Governor, the Bank said in August 2013 “it will be appropriate at a minimum to maintain the current stance of monetary policy at least until the LFS headline measure of the unemployment rate falls to 7%”. At that point, interest rates were 0.50% and the unemployment rate was 7.6%.

Just five months later, the unemployment rate had fallen below 7.0% and the Monetary Policy Committee (MPC) was nowhere near ready to raise rates so it tweaked its guidance. In the end, rates remained at 0.50% until July 2016 when they were cut after the EU Referendum. They were not raised above 0.50% until mid-2018, when the unemployment rate was 4% and five years after it fell below 7%! (See Chart 9.)

This led one member of the Treasury Committee to accuse Carney of behaving like an “unreliable boyfriend… one day hot, one day cold, and the people on the other side of the message are left not really knowing where they stand”.

Chart 9: Bank Rate & Unemployment Rate

Source: Refinitiv

The Bank will certainly use some form of words to signal that it intends to keep policy loose for a prolonged period. But we doubt this itself will provide much of a boost to economic activity.

Negative interest rates

Negative interest rates are more compelling. Since the GFC, the Bank has been against negative rates as it has always concluded the damage to the financial sector could outweigh any benefits to the broader economy. Indeed, in mid-March the current Bank Governor, Andrew Bailey, said they are “not something we are currently planning for or contemplating.”

But he seems to have had a change of heart since then. At the end of May he said “it is right that we consider what further options, such as cutting interest rates into unprecedented territory, might be available in the future.” That’s quite an about-face!

Clearly the Bank is trying to learn from other economies, such as Japan, the euro-zone and some Scandinavian economies, that have had negative interest rates for many years. The Swiss National Bank has gone the furthest by setting an interest rate of -1.0%. (See Chart 10.) And it is thought negative rates have had some positive effects. (See here.)

Chart 10: Policy Rates (%)

Source: Refinitiv

But there are two reasons why negative interest rates aren’t a magic bullet. First, they don’t provide the Bank with that much more ammunition because of what’s called the Effective Lower Bound (ELB). That is the lowest that rates can be before banks and households start to hold cash rather than keep their money in the bank, which could then destabilise the financial system and make monetary policy far less effective.

Academic studies suggest that the cost of holding cash, such as storage, insurance and transportation, are around 1%. That means the Bank could cause problems if it cut interest rates to -1.0%. Admittedly, the coronavirus crisis has decreased the use of cash and accelerated the use of electronic payments. And the ELB does evolve over time. A few years ago the Bank thought it was around 0.50%, but it has since cut rates to 0.10%. Even so, the Bank is unlikely to conclude that the ELB has plunged. So negative rates perhaps create extra policy space of about 100bps at the most.

Second, there are concerns that rate cuts below zero are less effective at stimulating demand than rate cuts above zero. The evidence suggests that negative interest rates successfully result in banks lowering their loan rates (i.e. their interest income). However, as banks are reluctant to charge depositors a fee for holding their cash because of the inevitable public backlash, banks’ deposit rates don’t tend to fall as far (i.e. their interest costs). This has certainly been the case in the euro-zone. (See Chart 11.) The result is that negative rates can reduce bank profitability and prompt banks to cut the supply of credit – the exact opposite of what the policy is trying to achieve!

Chart 11: ECB Policy Rate & Bank Deposit Rate (%)

Source: Refinitiv, Capital Economics

Such downsides are typically thought to be bigger in the UK than in other countries due to the size and complexity of the UK’s financial sector. None of this is insurmountable, though, and five points are worth making.

First, it’s worth noting that, unlike in the US where it is not even clear if negative rates are legal, there are no such legal challenges in the UK. Second, although the UK’s large financial sector could mean that any downsides are magnified compared to elsewhere, it is the retail banking sector that is influenced the most by negative rates. And the UK’s retail banking sector is similar in size to those in Europe.

Third, within the retail banking sector the UK’s building societies are more in the firing line as they are required by law to raise at least 50% of their funding from deposits. As such, it is thought that negative rates would hit their profitability by more as they would need to keep deposit rates higher than banks. But given that Nationwide, the largest building society, currently gets about two thirds of its funding from deposits there is some room for it to take more funds from the market. And there is a provision within the law that in exceptional circumstances building societies can source only 25% of their funding from deposits.

Fourth, the Bank of England can mitigate some of the costs to banks by providing them with cheap funding through the existing TFSME scheme. Since its inception in March, £11.9bn has been lent to banks through this scheme. As the Bank estimates that its capacity is around £200bn, there is still plenty of room for banks to reduce their funding costs by borrowing from the Bank. The Bank could make the terms even more favourable to help reinforce the pass-through of negative rates to the wider economy.

Fifth, the Bank could give commercial banks a break on the fees they would have to pay the central bank if rates were negative. At the moment, the Bank’s QE essentially forces the commercial banks to hold more reserves at the central bank, which currently earn interest at Bank Rate of 0.10%. If Bank Rate is cut below zero, then the commercial banks would have to pay the Bank a fee to hold their reserves. So the Bank would be forcing the banks to hold reserves with it and charging them for the service at the same time! That’s a nice little earner, but it would hardly prompt the banks to lend more.

The Bank of England could copy the ECB and Swedish Riksbank by splitting bank reserves into required reserves and excess reserves and only charging banks a fee on the latter. In fact, the Riksbank exempts 30 times banks’ required reserves from paying negative interest rates. There’s nothing stopping the Bank of England from doing something similar.

So in practice the Bank can introduce negative rates and take some steps to mitigate the potential pitfalls. But the Bank probably won’t be in a position to do this all overnight. And the bigger point is that negative rates don’t give the Bank a huge amount of extra ammunition to deploy and it’s probably going to be reluctant to use it. As such, if it does use negative interest rates it would probably test the water in baby steps so it can assess the consequences. So we wouldn’t rule out negative rates completely, but they are unlikely to contribute much to any further loosening in policy.

Quantitative easing

This means that QE will probably account for the bulk of the 250bps of extra support we think is necessary. The strength of QE is that its potential size is essentially unlimited, it can be implemented quickly and easily, and the Bank has experience in assessing its effects. (See Box 1.)

Box 1: The transmission mechanism of QE

QE operates through six main channels:

  1. Monetary policy signalling: Providing extra information about the outlook for interest rates. For example, by signalling the Bank’s determination to meet the 2% inflation target, asset purchases may lead the markets to expect policy rates to remain lower for longer.
  2. Portfolio rebalancing: As the Bank of England buys assets, this raises the money holdings of the sellers. Unless money is a perfect substitute for the assets they sold to the Bank, the sellers may rebalance their portfolios by buying other assets that are better substitutes. This raises the prices of typically riskier assets and lowers their yields. This then lowers borrowing costs for firms and households, and higher assets prices create a wealth effect too.
  3. Liquidity effects: Improving liquidity by encouraging trading and reducing risk premia on various assets.
  4. Exchange rate: Raising the price of domestic assets lowers their yield relative to overseas assets and weakens the exchange rate.
  5. Confidence and uncertainty: Improving the economic outlook and reducing the risk of bad outcomes. This may boost the confidence of consumers and their willingness to spend.
  6. Bank lending: Encouraging banks to extend more new loans than otherwise.

Admittedly, there are question marks over how effective more gilt purchases would be. While QE helped to push down gilt yields in 2009, 2011/12 and again in 2016, with 10-year yields currently not much more than 0.20% there is not much scope for yields to fall further. And although the signalling element is still important, as the market has begun to anticipate the use of QE there would be less of a “shock” factor than before. Finally, the high supply of gilts means that the private sector may essentially be selling gilts to the Bank and buying gilts from the government, which would reduce the effectiveness of the portfolio rebalancing channel as the private sector is just switching between two identical assets.

That is not to say that an expansion of QE would be completely ineffective. It could still signal that the Bank is willing to do whatever it can to hit the 2% inflation target. And while QE typically works best during times of financial stress, the policy could still lift confidence during the current uncertain period. And even if QE doesn’t reduce yields further, it may at least prevent yields from rising.

Based on the Bank’s estimate that in 2011/12 every £36.5bn of asset purchases was equivalent to a 25bps interest rate cut (which probably overstates the effectiveness now), a loosening in policy equivalent to a 250bps rate cut would require QE to be expanded by around £365bn. Given that the Bank can’t pretend to fine tune the economy exactly, a round number of around £350bn is more likely in practice.

Credit easing

It is possible that the Bank could put more emphasis on credit easing (i.e. purchasing assets that carry more risk than gilts). Admittedly, the Bank has already pledged to double its holdings of investment grade corporate bonds from companies that make a “material contribution to the economy” from £10bn to £20bn. And it’s possible that of the £350bn of extra QE that we think is required, investment grade corporate bonds could account for £10-20bn.

In addition, the CCFF is a form of credit easing as it involves the Bank buying 1-year commercial paper direct from businesses (the Treasury covers any losses). So far the Bank has bought £19bn of assets through the CCFF and there is no upper limit.

But the Bank could get a bigger bang for its buck by either buying a higher share of investment grade corporate bonds relative to its purchases of gilts or by purchasing riskier assets, such as non-investment grade bonds (“junk” bonds) or equities. The Bank of Japan pioneered the latter in 2011 with its purchase of equity Exchange Traded Funds (ETFs). Any of these options should lead to a larger portfolio rebalancing effect as investors are forced to seek riskier assets to replace the ones they have just sold to the Bank.

At the same time, as the corporate bond market is less liquid than the gilt market, the extra demand from the Bank would, at least initially, result in a bigger improvement in the functioning of the market. And given that corporate bonds are typically issued by larger firms, if those firms then borrow less from banks then that could increase the supply of credit to smaller firms.

However, while the size of the corporate bond market has doubled since the GFC, at just shy of £300bn it is still less than 20% the size of the gilt market (£1,718bn). (See Chart 12.) And as corporate bonds are traded more infrequently than gilts and are issued in relatively small sizes, there is a limit on how much the Bank could practically buy.

Chart 12: Value of Sterling Non-Financial Corporate Bonds in Issue (£bn)

Source: Merrill Lynch

Buying riskier assets like non-investment grade bonds or equities would also mean the Bank would have to shoulder the risk of losses. In 2018, losses made by the ECB on a bond issued by the South African retailer Steinhoff prompted a lot of criticism of the ECB’s policy. So at the very least, the Treasury would need to be willing to cover any such losses.

So while we suspect that the Bank will be willing to continue with its current credit easing policy of buying investment grade corporate bonds and commercial paper through the CCFF, we’re not convinced that it is about to take a big leap into buying riskier assets such as junk bonds or equities.

Yield curve control

In a more radical step, the Bank could ditch QE altogether and adopt Yield Curve Control (YCC), a policy adopted by the Bank of Japan in 2016. This involves the Bank setting a price for assets rather than setting a quantity it intends to buy. In the case of the Bank of Japan, since September 2016 it has said it will calibrate its asset purchases to keep the Japanese 10-year bond yield close to zero. (See Table 2.)

In theory, YCC could require unlimited purchases as if there is upward pressure on bond yields then the central bank has to buy as many assets as necessary to bring yields back down to the target rate. But if the target is deemed credible, then the central bank doesn’t have to buy any assets as its signal does all the work for it.

Table 2: Yield Curve Control in Action

Country

Date

Target

Aim

Australia

2020

3-year gov’t bond yield of 0.25%

Improve the functioning in the markets.

Japan

2016

10-year gov’t bond yield around 0%.

To stop yields falling below zero and reducing the profitability of insurance companies after the introduction of negative interest rates.

US

1942-51

3/8% on 3-mth gov’t bonds.

2½% on long-term bonds.

Ceilings for short- and long-term bond yields to keep government borrowing cheap during WW2.

Source: Capital Economics

This is exactly what happened to the Bank of Japan. Once it started YCC, it was able to purchase fewer assets. (See Chart 13.) In fact, the Bank of Japan didn’t really implement YCC to stop yields from rising. It did it to stop long-term yields from falling too far! That’s because it feared that falling long-term rates were reducing bank profitability and hindering the flow of credit.

Chart 13: Japan Bond Yields & Annual Change in Bank of Japan’s JGB Holdings

Sources: Refinitiv, Capital Economics

The Bank of England is certainly not in that situation now. But equally, with 10-year gilt yields currently very low at close to 0.20%, there is little need to set a target to keep yields where they are already. What’s more, if YCC led to fewer asset purchases that would arguably lead to the Bank’s policy actions boosting the economy by less as the portfolio rebalancing and bank lending channels would be less effective.

The one situation in which YCC could become really appealing is if in response to the explosion in public debt caused by the government’s response to the coronavirus a fear of debt sustainability caused gilt yields to start rising significantly. At that point, the Bank could use YCC to keep yields down and allow the government to access cheap borrowing. That would replicate the situation in the US during and after World War 2 and would be a form of financial repression. This is certainly a risk worth considering, but it is not something we are forecasting.

Helicopter money

The most extreme option would be if the Bank co-ordinated with the Treasury on a money-financed fiscal expansion. This is the helicopter drop in its purest form as the government would ask the Bank to print some money and the government would hand it out.

On the face of it, the Bank already appears to be doing something similar. The Office for Budget Responsibility has estimated that the government’s direct tax breaks and spending pledges during the coronavirus crisis amount to £123bn and we think the eventual cost will be even higher. (See here.) At the same time, the Bank is on track to complete its £200bn of asset purchases (£190bn of those gilts) by early July. Essentially, the government is increasing its spending knowing that the Bank will print money to pay for it.

However, the two key defining characteristics of monetary financing are not present. First, the Bank is not funding the government directly. Instead, it is buying government bonds on the secondary market. That means the market is setting the price not the government. Second, the Bank’s QE is intended to be temporary and the primary aim is not to fund the government’s spending but to meet the Bank’s inflation target. A pure helicopter drop requires the creation of the extra money to be permanent. (See here.)

Of course, this doesn’t mean that the situation can’t change. And you could argue that the UK is better placed than some other economies to reap the benefits of monetary financing without the possible cost of runaway inflation. After all, the government respects the independence of the Bank of England and compared to many economies around the world the UK is hardly a high inflation economy. Even, a lot would need to change, perhaps including an erosion of the Bank’s independence, for the Bank to embrace pure monetary financing.

What and when?

The main message is that it would be unwise to rule out any of these tools. After all, desperate times call for desperate measures! And the longer the crisis lasts and the longer more policy support is needed, the more experimental the Bank will become. If that’s not the case during the current crisis, then it will surely be the case during the next one as by then the Bank may not have replenished its policy arsenal.

As such, negative rates, credit easing and YCC are all possible at some point. But for now, we think that QE will do most of the heavy lifting, mostly focused on purchases of gilts but with some purchases of investment grade corporate bonds.

The size and timing of any future action is even more uncertain than whether there will be any more action in the first place. But history provides three useful guides that our forecasts are based on.

First, in previous rounds of QE the Bank’s initial announcement has always been the largest and subsequent increases in QE have always been smaller. (See Chart 14.)

Chart 14: QE Announcements (£bn)

Sources: Bank of England, Capital Economics

Second, the Bank tends to take its time and waits for its current round of QE to be more or less completed before announcing a new one. The gap between the first and last announcements of QE in 2009 and 2011/12 was eight and nine months respectively.

Third, the Bank’s actions aren’t confined to the immediate period of crisis. The yellow dotted lines on Chart 15 show that the Bank was still announcing additional QE in 2009 even though GDP was growing again.

Chart 15: GDP (%q/q) & BoE Policy Action

Sources: Refinitiv, Bank of England, Capital Economics

This time, we think the MPC will announce a further £100bn of QE at its meetings in June, August and November as well as a further £50bn in February next year. (See Table 3.) When taken together with the £200bn announced in March, that would take the Bank’s total stock of QE assets to £995bn. That would be almost a 55% increase from the stock of £645bn before the coronavirus crisis and would leave the Bank holding almost £1trn in assets.

The risks

If anything, there are two reasons why the Bank may end up loosening policy by more than we expect. First, the risk is that even our relatively downbeat economic forecasts prove to be too pessimistic and the economy requires more than the equivalent of a 250bps rate cut.

Second, MPC member Michael Saunders advocated in a speech in May a “risk management” approach that “implies…if easing is needed, it should occur promptly” and “it is better to err on the side of somewhat too much stimulus rather than too little”. If the rest of the MPC agrees, the Bank could either announce something similar to the total QE package we have assumed sooner than we expect, or it could increase QE in fewer bigger steps, or the total increase in QE could be bigger and/or it could test out some of the other tools we have discussed.

Table 3: MPC Meeting Outcomes & Forecast

Date

Rates

QE

30th Jan. 2020*

0.75%

11th Mar. 2020 (Emergency meeting)

0.25%

19th Mar. 2020 (Emergency meeting)

0.10%

+£200bn

26th Mar. 2020

0.10%

7th May 2020*

0.10%

18th Jun. 2020

0.10%

+£100bn

6th Aug. 2020*

0.10%

+£100bn

17th Sep. 2020

0.10%

5th Nov. 2020*

0.10%

+£100bn

17th Dec. 2020

0.10%

4th Feb. 2021*

0.10%

+£50bn

18th Mar. 2021

0.10%

6th May 2021*

0.10%

24th Jun. 2021

0.10%

5th Aug. 2021*

0.10%

23rd Sep. 2021

0.10%

4th Nov. 2021*

0.10%

16th Dec. 2021

0.10%

Sources: Bank of England, Capital Economics. *Denotes publication of Monetary Policy Report.

Will any of this actually work?

Even if the Bank does loosen policy by as much as we think is required, there is still a big risk that it won’t be enough to get the economy back on track and raise inflation to the 2% target. After all, we know that the effectiveness of monetary policy diminishes once interest rates have been reduced close to zero and that subsequent rounds of QE tend to be less potent than the initial burst. That suggests even if our economic forecasts prove to be broadly right, an extra £350bn of asset purchases may still not be enough.

The implication is that monetary policy won’t be able to solve all the economy’s problems on its own. There has already been a vast loosening in fiscal policy aimed at keeping businesses and households afloat during the coronavirus lockdown. But at some point in the coming months, fiscal policy will need to change tack and create incentives for businesses and households to spend. So both monetary and fiscal policy will be required to work in tandem and the diminishing effectiveness of monetary policy means that fiscal policy will have to do most of the heavy lifting.

When will monetary policy be “normalised”?

So a period of ultra-loose monetary and fiscal policy has already begun and we suspect this is more likely to last for many years rather than one or two, not least because of the interaction between monetary and fiscal policy.

With the government likely to be running chunky deficits for the next five years or longer, there is a clear incentive for loose monetary policy to keep borrowing costs low so that government debt doesn’t get out of control. If it did, then the Bank would find it hard to meet its inflation target.

We doubt that the Bank will raise interest rates above 0.10% in the next five years. That would mean interest rates stay at their current level or below for roughly two years longer than the financial markets currently expect. (See Chart 16.)

Chart 16: Forecasts for Bank Rate (%)

Sources: Refinitiv, Capital Economics

If anything, rates could stay at that level or below for even longer. After all, interest rates were stuck at 0.50% for seven years after the GFC before they were cut in 2016. They only rose above 0.50% nine years after they were originally reduced to that level. (See Table A1 in the Appendix.)

And it may be much longer, perhaps even 10 years, before the Bank starts to unwind QE. In fact, if the experience after the GFC is anything to go by, the Bank may not be able to reduce the stock of QE at all before the next crisis!

This assumes that the Bank sticks to the plan it put in place after the GFC, namely that when it gets round to contemplating tighter monetary policy the first step would be to raise interest rates long before unwinding QE. That’s to ensure that it has enough space to cut rates materially in response to future shocks.

In the years after the GFC, the Bank suggested that it would want to first raise interest rates to around 2%, which given that it thought the Effective Lower Bound was 0.5% meant it was aiming to have the ability to cut rates by around 150bps. After it subsequently reduced its estimate of the Effective Lower Bound, in 2018 the Bank said it wouldn’t start unwinding QE until interest rates had been raised to 1.5%. If the Bank were to lower its estimate of the Effective Lower Bound again, then it may start to think about unwinding QE when interest rates have been raised to between 1.0% and 1.5%.

Either way, given that businesses are going to exit this crisis with more debt than ever, when the Bank does finally get round to raising interest rates it will do so very slowly and cautiously as any given rate rise would be more effective at reining in demand. So it will be a long time before interest rates get to 1.0% or above.

Overall, we think that one of the legacies of the coronavirus crisis will be five years of interest rates of no higher than 0.10% and a decade of ultra-loose monetary policy.


Appendix

Table A1: Bank of England Monetary Policy Timeline

MPC Meeting Date

Interest Rates

QE

From

To

Change

Quantity

Cumulative

Timespan

%

%

bps

cum.*

£bn

Gov’t

Corp.

% of GDP

£bn

Gov’t

Corp.

% of GDP

Months

Global Financial Crisis

Dec-07

5.75

5.50

-25

-25

Feb-08

5.50

5.25

-25

-50

Apr-08

5.25

5.00

-25

-75

Oct-08

5.00

4.50

-50

-125

Nov-08

4.50

3.00

-150

-275

Dec-08

3.00

2.00

-100

-375

Jan-09

2.00

1.50

-50

-425

Feb-09

1.50

1.00

-50

-475

Mar-09

1.00

0.50

-50

-525

75

75

0

4.9

75

75

0

4.9

3

May-09

50

50

0

3.2

125

125

0

8.1

3

Aug-09

50

50

0

3.2

175

175

0

11.2

3

Nov-09

25

25

0

1.6

200

200

0

12.9

3

Euro-zone Debt Crisis

Oct-11

75

75

0

4.5

275

275

0

16.4

4

Feb-12

50

50

0

3.0

325

325

0

19.3

4

Jul-12

50

50

0

2.9

375

375

0

21.8

4

UK Votes to Leave the EU

Aug-16

0.50

0.25

-25

-25

70

60

10

3.5

445

435

10

22.2

6 gov’t
18 corp.

Nov-17

0.25

0.50

+25

0

Aug-18

0.50

0.75

+25

+25

Coronavirus Pandemic

Mar-20+

0.75

0.25

-50

-50

Mar-20+

0.25

0.10

-15

-65

200

190

10

9.1

645

625

20

29.3

ASAP^

Sources: Bank of England, Capital Economics *Cumulative change during period. +Emergency MPC meetings on 11th March and 19th March. ^As soon as operationally possible.


Paul Dales, Chief UK Economist, +44 7939 609 818, paul.dales@capitaleconomics.com
Ruth Gregory, Senior UK Economist, +44 7747 466 451, ruth.gregory@capitaleconomics.com
Thomas Pugh, UK Economist, +44 7568 378 042, thomas.pugh@capitaleconomics.com
Andrew Wishart, UK Economist, +44 7427 682 411, andrew.wishart@capitaleconomics.com
James Yeatman, Research Economist, +44 20 7808 4694, james.yeatman@capitaleconomics.com