How big should the central bank balance sheet be?

After the Global Financial Crisis of 2007-9, the world’s largest central banks undertook new policies of quantitative easing. This included the US Federal Reserve, the ECB, the Bank of Japan and the Bank of England. 

The asset-buying spree was intended primarily to counter deflationary tendencies. Although there were differences in which assets were bought, each central bank expanded the supply of narrow money.

That liquidity injection which drove up financial asset prices, reduced interest rates across the yield curve, and squeezed credit and liquidity spreads. 

With the exception of Japan, inflation was largely back to target by early 2017. Asset purchases were largely slowed or stopped and the United States even began a reversal in 2018.

But in 2020 the pandemic struck, and the expansions started anew.

By the end of 2021, each central bank had increased the size of its balance sheet to between five and ten times their end-2006 levels.

Inflationary pressures started to build during 2021, exacerbated by the consequences of the Russian invasion of Ukraine.

Monetary policy was suddenly about-turn, trying to bring inflation down from levels not experienced for decades.

With inflation, not deflation the main challenge, what happens to expanded central bank balance sheets.

Does quantitative easing get unwound completely, partly, or not at all? How big should the central bank balance sheet be?

The text book answer is if the central bank wants to implement monetary policy by setting an interest rate - the price of central bank money – then the quantity of that money has to be consistent with the demand for it at the set price. 

The total demand for central bank money is driven by the desired level of commercial bank reserve accounts held at the central bank.

The central bank controls the supply by expanding or contracting its assets – a mixture of purchased assets held outright and its secured loans to the banking system.

The demand for reserve accounts by commercial banks appears to have shifted. Demand has increased in part because of new prudential requirements for banks to hold more liquid assets.

And, many smaller banks have built business models on attracting the deposits created by the central bank cash injection. This is particularly common in the US.

The demand for reserve accounts is not easy to model, and the interest rate effect is not well-determined.

Nobody really knows what level of reserves is going to be consistent with any particular level of interest rates.

Less than current levels, certainly, and much higher than it was before the Global Financial Crisis. However, that leaves a large range of uncertainty.

Why do central banks want to reduce their balance sheets?

Central banks will want to reduce their balance sheets to the lowest level consistent with monetary policy. Why? Firstly, because it would then be easier to expand them should a new liquidity crisis emerge. 

Secondly, because it reduces risk - market and credit risk to the central banks’ assets, and operational risk from large operations.

Thirdly because quantitative easing has clearly had distortionary effects on markets which could be usefully avoided.

And fourthly because of the fiscal consequences of quantitative easing, which will have made many central bankers nervous.

The process will be a slow contraction. As liquidity is withdrawn, some (probably smaller) commercial banks will become short. Other banks (and the system as a whole) will remain long.

With inter-bank markets less efficient than they were, some interest rate volatility could emerge. This was seen when the Fed first allowed its balance sheet to shrink in September 2019. 

Central banks will meet this risk by expanding their lending operations at the same time as reducing their outright asset holdings, in order to make sure that liquidity reaches all parts.

Nobody knows exactly where the normalisation process will end, but it’s going to be interesting finding out.

If I were to guess, I would suggest maybe half of the peak levels for the US, Japan, and the Euro area. And perhaps pre-Brexit referendum levels for the UK area. And perhaps pre-Brexit referendum levels for the UK.

This article was first published by UK Finance.

Further reading:

Why would central banks what to issue digital currencies?

What does a digital pound mean for commercial banks?

Is ring-fencing UK banks really necessary?

Why central banks need to understand behavioural finance

Paul Fisher is an Honorary Professor at Warwick Business School and a former member of the Bank of England's Monetary Policy Committee. He teaches on the Global Central Banking and Financial Regulation qualifications.

Discover more about banking, finance and markets by subscribing to our free Core Insights newsletter.