It might be a distant memory now, but back in 2009 when financial markets were crashing and queues were forming outside banks, the UK Government along with the Bank of England hatched an emergency plan to stop the financial system from completely collapsing.

Quantitative easing or QE - the buying up of large quantities of bonds by central banks to stimulate economic activity and inflation - was thrust into the popular lexicon. But far from being an emergency measure, this so-called unconventional monetary policy is still being used by central banks around the world today.

After the UK’s EU referendum in 2016 the Bank of England hoovered up £70 billion of bonds to address uncertainty over Brexit. While the European Central Bank (ECB), after finally ending its initial QE programme in 2018, then announced it would be starting it back up again in 2019 with the Eurozone economy still floundering with low inflation and poor growth. And in Japan QE has become a regular occurrence since 2010.

But how do central banks measure the effect of QE? The majority of central banks in the developed world use inflation-targeting to stabilise economic growth and inflation, and the way they normally do this is by controlling the base interest rate.

In the case of the Bank of England, this is the bank rate, whereas in Brazil it is the SELIC rate. By controlling this short-term policy rate, central banks aim to control the cost of borrowing for business and households via the effect that the policy rate has on banks’ lending rates.

To measure the causal effects of base rate setting we need to compare the counter-factual situation that the economy will keep moving as usual with the one that the economy was disturbed by a monetary policy change.

To solve this problem, economists rely on the work of Chris Sims, who won the Nobel Prize in Economic Sciences in 2011. The idea is to measure the effects of monetary policy shocks, that is, changes in the policy rate that are not a result of previous changes in the economy.

Professor Sims devised Vector Autoregressive (VAR) models to measure the effects of monetary policy in an economy. These empirical models allow us to say how much a change of, say, 25 basis points in the policy rate will have on GDP growth and inflation and how long in months the effect would last.

The calculations can be done very quickly thanks to the VAR algorithms and it is something we do with students on the MSc Global Central Banking and Financial Regulation.

Download the brochure on MSc Global Central Banking and Financial Regulation

But the financial crisis of 2007-08 ushered in a new problem for inflation-targeters. By 2009 interest rates had hit the ‘zero-lower bound’ as central banks tried to reboot the economy. The UK’s policy rate was reduced to 0.5 per cent, where it stayed until 2018 when it was raised to 0.75 per cent, while the US Federal Reserve reduced it’s to 0.25 per cent and the ECB reduced its policy rate to one per cent in 2009 and it reached negative levels at -0.5 per cent in 2019.

With nowhere to go QE was born. Instead of manipulating the bank rate, the central bank buys long-term bonds from financial institutions, such as pension funds, with the aim of reducing the long-term interest rate. The idea is that buying up bonds pushes bond prices up and the yields down and so eases the cost of borrowing.

It has led to accusations of central banks “printing money” and inflating the money markets, but not affecting the ‘real economy’. And certainly it has created a new challenge for economists. How do we measure the effect of QE, with the standard Sims calculations now largely redundant?

In an effort to extract all the many variables that can affect an economy, economists designed ‘event studies’ to see how the central bank announcements of a QE policy affected bond yields. These event studies aimed to measure the causal effects of QE policy in the financial markets by measuring how unexpected announcements changed yields. 

Central banks actually look at a very short window with minute-by-minute data from financial markets after the announcement, so before they have even started buying bonds. And in the US the Federal Reserve is looking at the futures market on Government bonds to try to measure the impact of the announcement of QE.

The reason they need a very short window is to eliminate confounding effects and isolate a causal effect, as the only new information is the QE announcement, so no other news would have affected the market. This, though, assumes that the market is efficient and captures all information available at each point in time.

Those measurements are then included in the VAR model to predict what will be the likely effect of any new QE on GDP and inflation.

QE does certainly have an effect on financial markets and there is some research showing evidence that it increases inflation and GDP, with one paper finding a 0.3 per cent rise in GDP and a similar impact on the Consumer Price Index, but there is some controversy around this and the impact on employment is doubtful at the moment.

The controversy reflects the difficulty in measuring the effect of QE and research is continuing into this area, especially as, a decade on, it looks like far from being an emergency measure, QE is here to stay.

Find out more about banking regulation on the MSc Global Central Banking and Financial Regulation or download a brochure.

Ana Galvao is Professor of Economic Modelling and Forectasing and teaches Monetary Policy and Monetary Analysis on the MSc Global Banking and Financial Regulation, delivered in partnership with the Bank of England.

For more articles like this download Core magazine here.