Stefano Fugazzi (ABC Economics) – Hereafter a synopsis of a research paper by the Bank of England (BoE) on the macroeconomic effects of Quantitative Easing (QE) and the impact of unconventional monetary policies on Gross Domestic Product (GDP). The BoE research unit found that the second round of the QE purchases during 2011–12 boosted GDP in the United Kingdom by around 0.5%–0.8%. Its effect on inflation was also broadly positive reaching around 0.6 percentage points, at its peak.
The pronounced global financial crisis in late 2008 led to a severe economic downturn on a scale notseen since the Great Depression of the 1930s. The fiscal and monetary authorities of many countries responded with a variety of conventional and less conventional measures aimed at mitigating the effects on financial stability and the real economy. In the aftermath of the crisis, the Bank of England (BoE), like many other central banks, introduced a number of innovative policy measures to loosen monetary conditions, as the scope for conventional policy rate reductions became increasingly constrained by the fact that interest rates were already close to their lower bound. These measures included enhanced liquidity support, actions to deal with dysfunctional financial markets and large-scale asset purchases.
QE, was initially announced in March 2009 at the height of the global crisis, with the first round of £200 billion of purchases, predominantly of UK government securities (gilts), concluding in January 2010. Our focus in this paper, however, is on the second round of purchases between October 2011 and June 2012, when the Bank of England purchased £175 billion of gilts, about 11% of nominal GDP, in response to concerns that the euro area sovereign crisis would lead to UK CPI inflation undershooting its 2% target.
A large and growing literature has developed over recent years that tries to evaluate the impact of the unconventional monetary policies adopted by central banks during the global financial crisis.1 Most of this literature has focused on the Fed’s various policy measures, and particularly its large-scale asset purchase programmes, and their impact on financial markets and macroeconomy. A large literature has also grown up around the ECB’s non-standard measures. In relative terms, much less has been written on the BoE’s unconventional monetary policies, with most of the published studies to date focusing on the financial market impact of the first phase of QE during March 2009 to January 2010.
This BoE paper examines the macro economic impact of QE1 during March 2009 to January 2010 using several time-varying VAR models of the economy to construct counterfactual simulations, assuming that the impact of QE came through a reduction in longer-term interest rates.
QE was primarily expected to affect GDP and inflation by reducing gilt yields and boosting other asset prices, thereby reducing the costs of borrowing, and boosting expenditure through wealth effects. This would occur mainly through a portfolio balance channel. Agents in the non-bank private sector would prefer to reinvest the newly created bank deposits into riskier assets such as corporate bonds and equities, absent a change in the price of those assets. This would lead to portfolio re-balancing and asset price changes, due to the imperfect substitutability of money and securities. Asset prices might also be boosted through a signalling channel that policy rates would remain lower for longer and also through improvements in market liquidity. Other secondary channels were conceived as possible, including a bank lending channel, although the Monetary Policy Committee (MPC) always downplayed the likely importance of the latter given that banks were in the process of de-leveraging.
In general, QE was intended to work by circumventing the banking sector. QE would in the first instance benefit the owners of assets, and businesses who could issue debt or equity in capital markets.
The following exhibit shows the announcement reactions over both 1 and 2 day windows and also extending the dataset to include the policy announcements associated with QE2. Summing over the QE1 events using the longer 2-day window suggests that overall 5-year yields fell by around 55 basis points and 10-year yields fell by 80 basis points. The reactions to the QE2 announcements are much smaller, however, particularly for longer maturities. In considering the impact of later purchases and particularly QE2, there are reasons for thinking that event study methods might be less useful, as financial markets may have become more familiar with the use of QE (i.e. the Bank’s reaction function) and therefore been better able to anticipate its use. At the same time there are reasons why it may not be appropriate to quantify QE2 on the basis of QE1 event studies, as the importance of some of the channels through which QE works may have been weaker in QE2 relative to QE1 (e.g. any impact through signalling might plausibly have been greater when the new policy was first announced than when it was extended).
The following table illustrates the effects of QE2 on GDP and annual inflation under the scenario that the yield spread was reduced by 45bps. The exercises suggest QE2 had a significant effect on demand. GDP rises 0.6 % by the end of the period under consideration. The inflation response has an inverted U shape, and is initially increasingly positive and significant, peaking at 0.6 pp. Then, it declines to about 0.25pp. These estimates can be contrasted with the effects of QE1 where the first monetary easing programme was found to have had a positive impact on GDP of about 1.5% and to inflation of about 1.25pp. Given our finding that QE2 had a smaller effect on spreads compared to QE1, these differences appear intuitive.