Media Briefings

EFFECTIVE PROGRAMMES OF ‘QUANTITATIVE EASING’: THE IMPORTANCE OF DURATION AND FORWARD GUIDANCE

  • Published Date: November 2012

The length of central banks’ ‘quantitative easing’ (QE) programmes and how they are communicated to the markets could be at least as important for their effectiveness as the scale of asset purchases. That is one of the conclusions of research by San Francisco Fed economists Jens Christensen and Glenn Rudebusch, published in the November 2012 issue of the Economic Journal.

Comparing the QE programmes in the UK and the United States, the study finds notable differences in how they affect yields on government bonds. These different outcomes seem to reflect differences in the structures of the programmes and how they are communicated. US programmes tend to be of longer duration; they also include some forward guidance about near-term prospects for the Fed’s policy rates.

The findings of this research suggest that relying more heavily on explicit forward guidance about the future path of policy rates and less on asset purchases could prove to be a cost-efficient way to obtain similar effects. Such an approach would also involve less risk for the central bank when it is time to exit its unconventional monetary policies and to normalise the size of its balance sheet.

For several years, the traditional monetary policy tool, a targeted overnight rate, has been at its effective lower bound in both the UK and the United States. To provide additional stimulus, the policy rate setting committees in both countries have purchased large quantities of long-term assets (also known as QE), in an attempt to lower long-term interest rates.

Even though the positive effect of such purchases appears to be well established in the literature of academic research, it is still poorly understood how these policies work and how QE programmes should be managed to be most effective.

In general, QE is thought to operate through one of two channels. In one view, QE causes yields to decline by changing expectations for future monetary policy through a signalling effect. In other words, the asset purchases signal that the target rate will remain low for longer than previously anticipated.

In an alternative interpretation, the asset purchases lead to yield declines by reducing risk premiums through what is called a ‘portfolio re-balance’ effect, as investors are forced to increase their demand for riskier assets.

Importantly, the two channels would imply rather different recommendations for monetary policy. At one extreme, if the effect of QE is solely a result of signalling, no purchases would be needed: only explicit forward guidance for future policy would be required.

At the other extreme, if the effect of QE is all associated with reductions in risk premiums, the size and composition of the asset purchases would be critical to their success.

This research analyses the response of UK and US government bond yields to central bank announcements about the QE programmes in each country. To distinguish between the signalling effect and the portfolio re-balance effect, the authors employ a dynamic term structure modelling approach that allows them to make a real-time decomposition of the yield changes into separate changes in policy expectations and risk premiums.

Surprisingly, the study finds notable differences in how the two countries' QE programmes affected yields despite many shared similarities. In the United States, about 60% of the yield declines reflect lower expectations for future monetary policy, while declines in risk premiums account for at most 40% of the effect.

In contrast, in the UK, all yield declines are associated with declines in risk premiums as expectations about future monetary policy are estimated to have firmed slightly, on net, in response to announcements about the UK’s QE programme.

To explain the differences in the response patterns to QE announcements, the authors point to differences in both programme structure and communication strategy. In the United States, the QE programmes have tended to be of longer duration – six to ten months compared with a typical duration of three months in the UK. The US QE programmes have also included some form of forward guidance about the traditional policy rate in the near term, which is absent in the UK.

As a consequence, the findings seem to indicate that programme length and communication management could be at least as important for the effectiveness of QE policies as the actual purchase amounts.

In practical terms, to rely more heavily on explicit forward guidance and less on asset purchases could prove to be a cost-efficient way to obtain a similar effect, but at less risk to the central bank when it is time to exit the unconventional policies and to normalise the size of its balance sheet.

ENDS

Notes for editors: ‘The Response of Interest Rates to US and UK Quantitative Easing’ by Jens Christensen and Glenn Rudebusch is published in the November 2012 issue of the Economic Journal.

Jens Christensen and Glenn Rudebusch are both at the Federal Reserve Bank of San Francisco.

For further information: contact Jens Christensen on +1-415-974-3115 (email: jens.christensen@sf.frb.org); Glenn Rudebusch on +1-415-974-3173 (email: glenn.rudebusch@sf.frb.org); or Romesh Vaitilingam on +44-7768-661095 (email: romesh@vaitilingam.com).