This op-ed was first published on CapX on Friday 21st June.
How the fiscal sums add up is an ongoing feature of the policy and political debate. Into this arena has stepped the issue of whether interest should be paid on the reserves of banks at the Bank of England (BOE). As it stands, this is costing us a huge amount of money. But to assess exactly how much could be saved, we first need to establish how we got here.
In the aftermath of the 2008 global financial crisis, interest rates fell close to zero and the BOE created money through quantitative easing (QE). QE’s goal was to boost the economy by lowering long-term interest, with the BOE buying government bonds (gilts) and financing this by creating money through the issuance of central bank reserves. It would buy government bonds in the secondary market from banks, crediting their account at the BOE: bank reserves. The BOE’s balance sheet would rise, as it held government bonds. In turn, banks would be paid interest at the policy rate on their bank reserves.
QE exchanged fixed rate debt for floating rate debt and initially made a profit. This was because the policy rate paid to banks was close to zero and the BOE was receiving a steady stream of income from the bonds it held. Importantly though, even though such operations were carried out by other western central banks, the UK took a different approach, with the Treasury indemnifying the BOE’s activities. Thus, if QE made money, the profits went to the Treasury while the taxpayer would pick up any losses.
At its peak, the BOE became the biggest holder of gilts. As QE reached £895bn, £875bn was held in gilts. Now its holdings are around £700bn as the BOE has been selling gilts through quantitative tightening (QT). As interest rates are now far higher, the taxpayer has a huge bill. The BOE is paying the policy rate of 5.25% to commercial banks on their reserves, and this dwarfs the return it receives on its gilts.
There are three options and the decision is a fiscal one, made by the government.
Option one is the status quo, continuing to pay commercial banks interest on their reserves at the BOE. This cost will remain sizable, but it would decline if interest rates fell. While interest rates look set to fall from 5.25%, they may then settle at a higher level than pre-pandemic, perhaps around 4%.
Another way the cost might fall is if the amount of commercial bank reserves decreases, and this is the counterpart to the BOE continuing to sell gilts through QT. Bizarrely, the BOE views this as a purely technical exercise. But it is not. It is monetary tightening at a time when inflation is falling, as it keeps yields high. Also, the BOE is ensuring a bigger loss to the taxpayer by selling gilts now, when yields are higher and their price is lower. The BOE should reduce gilt sales or even hold them to maturity.
Option two, which is my preference, is a move to a tiered system, where the banks are paid zero interest on some of their reserves and receive the current base rate on the rest. This is the approach used by the European Central Bank. It would save the taxpayer money, with the saving heavily influenced by where the tiers are set, as well as what happens to interest rates.
By paying its policy rate on reserves, as now, the BOE ensures that its decision on the level of policy rates is transmitted through the financial system. This still works in a tiered system, as it is the rate that is paid on marginal reserves that matters. Technically though, if too large a tier of reserves is paid zero, this may undermine that transmission. In turn, the BOE could freeze those reserves that are tiered to stop banks doing this. But that would create another challenge as banks count such reserves as high quality liquidity assets (HQLA) to meet regulatory requirements, and if frozen, it would be hard to claim they are HQLA.
Option three would be to pay zero interest on all the reserves, saving the taxpayer a large sum. But this would necessitate other market measures to make it work. With the BOE paying zero interest on reserves, then the fear is those banks will be able to push their rates down to zero, undermining the transmission of monetary policy.
This is valid, but let’s remember this is how life used to work in the City, and that it was changed not because the previous system was ineffective but because the way we have evolved to now is seen as more efficient. For instance, with option three there might be greater volatility in overnight rates, and thus if market rates are pushed below the BOE’s policy rate, there would need to be a draining of liquidity.
Banks would view any change to the status quo as a tax on them. A concern is that a tax on banks would see customers suffer or business go into less regulated spaces. Others have said there is a need to differentiate between a tax that would affect bankers (and how they reward themselves and their shareholders), as opposed to what impacts banking (and what feeds through into the wider economy). That’s key. After all, as the Bank of England noted in 2019 that the funding gap faced by small and medium-sized firms was £22bn.
It has also been argued that any change to the terms would constitute a default, which is an exaggeration, but it would be equivalent to a changing of the original terms and thus could undermine the credibility of future policies. The trouble though is that the scale of QE became too high and that the indemnity was a wrong commitment, unique to the UK, and not followed by other countries. Also, the financial backdrop has changed considerably, meriting a change.
Fundamentally, paying interest on bank reserves is about getting the balance right between saving the taxpayer money and ensuring monetary policy operates efficiently.