Media Briefings


  • Published Date: November 2012

The large-scale asset purchase programmes (LSAPs) that several central banks have implemented to deal with the aftermath of the financial crisis are likely to have moderate effects on real GDP and inflation. That is one of the conclusions of research by Han Chen, Vasco Cúrdia and Andrea Ferrero, published in the November 2012 issue of the Economic Journal.

Their study also finds that the effects are greatly dependent on the central banks’ commitment to keep their policy interest rates at the zero lower bound. What’s more, the effects are somewhat weaker and subject to more uncertainty than a 25 basis points reduction in the Federal Reserve’s policy rate, the federal funds rate.

The researchers develop a framework that highlights one channel through which LSAP programmes can affect the economy. The key feature that makes asset purchases relevant in their analysis is a form of financial market segmentation known in the research literature as ‘preferred habitat’.

According to this theory, an asset purchase programme that successfully reduces the yield on long-term bonds should tilt the profile of consumption towards the present and stimulate investment. This will have positive consequences for both output and inflation.

The study embeds this feature in a model commonly used in academic research and policy circles, which is known as a ‘dynamic stochastic general equilibrium’ model. It then uses the model to estimate the effects of the Fed’s LSAP II, in which $600 billion of assets were purchased.

The authors’ estimates suggest that the LSAP II programme has only moderate consequences for macroeconomic variables. Real GDP growth increases by 0.1% on impact and remains above trend for about two years. The effect on the level of real GDP peaks at 0.1% after six quarters and dissipates very gradually. The programme leads to an extremely modest increase in inflation (three basis points).

The interaction between interest rate policy and asset purchases plays a crucial role. In their main simulations, the authors assume that the central bank commits to keeping the interest rate at the zero lower bound for four quarters (consistent with market expectations at the time of the LSAP II announcement).

In an alternative scenario, in which the commitment to keep the interest rate at the zero lower bound is completely absent, the impact on output and inflation would be about halved. This exercise suggests that the effects of LSAP programmes greatly depend on expectations about interest rate policy.

The authors also compare the effects of LSAPs on real GDP and inflation with the typical response to a more standard 25 basis point surprise cut in the federal funds rate. They find that the impact of LSAP II on real GDP is about half of what the shock to the federal funds rate would generate, while the implications for inflation are comparable. The uncertainty surrounding the response of GDP growth is much smaller in the case of a federal funds rate shock.

The main lesson of this analysis is that asset purchase programmes like LSAP II are likely to have moderate effects on macroeconomic variables and the effects are dependent on the commitment to keep the interest rate at zero. The key reason is that there is little evidence in the data in favour of bond market segmentation, the feature that rationalises this type of intervention.

The authors discuss whether the financial crisis may have increased the degree of segmentation in bond markets, which would augment the effects of these policies. At the same time, other factors, such as uncertainty about the sensitivity of the risk premium to the volume of purchases, point in the opposite direction.

On balance, the authors argue that their results provide a reasonable benchmark for thinking about the costs and benefits of LSAPs going forward.


Notes for editors: ‘The Macroeconomic Effects of Large-Scale Asset Purchase Programmes’ by Han Chen, Vasco Cúrdia and Andrea Ferrero is published in the November 2012 issue of the Economic Journal.

Han Chen is at the University of Pennsylvania. Vasco Cúrdia is at the Federal Reserve Bank of San Francisco. Andrea Ferrero is at the Federal Reserve Bank of New York.

For further information: contact Vasco Cúrdia on +1-415-977-3624 (; Andrea Ferrero on +1-212-720-8903 (; or Romesh Vaitilingam on +44-7768-661095 (email: