The ECB and its Watchers XVI Conference
Speech by Mario Draghi, President of the ECB,
Frankfurt am Main, 11 March 2015
In January, the ECB decided to expand its asset purchase programme to include government bonds after it became clear that there was a need for more monetary stimulus. Asset purchases are unconventional, but not unorthodox, and they have been part of the ECB’s toolkit from the start. By deploying this tool, the ECB underlined its ability and determination to stabilise euro area inflation in line with its objective.
The impact of the programme and the ECB’s previous monetary policy measures is visible: Bank lending rates to companies started to decline in the third quarter of last year, market-based measures of inflation expectations have reacted positively to the ECB’s balance sheet expansion over recent months, and euro area long-term sovereign yields have fallen – in spite of the renewed crisis in Greece. This suggests that the asset purchase programme may be shielding other euro area countries from contagion, which also helps the ECB achieve its monetary policy goals across the euro area.
The euro area economy grew more than expected in the fourth quarter and unemployment fell to its lowest level since August 2012 in January. While this cannot exclusively be attributed to the ECB’s monetary policy, it certainly supports the recovery. Even though inflation is expected to remain very low or negative in the months ahead mainly due to the sharp drop in oil prices, it is expected to move closer the ECB’s policy target over the coming years to reach 1.8 per cent in 2017 - conditional on the full implementation of all policy measures.
The beneficial impact of the ECB’s asset purchases on financing conditions will increase the benefits of governments’ structural reforms, rather than reducing incentives for reforms. Firms will be encouraged to increase investment, bringing forward the economic recovery.
Ladies and Gentlemen,
Since the last Watchers’ conference in March 2014, we have reached deeper in our monetary policy tool box. The means have changed quite substantially, as dictated by circumstances, but the end has not. To be able to deliver on our medium term price stability mandate, we needed to adapt and expand our monetary policy tools.
I would like to open this conference by elaborating on what has changed, how it has changed, and how this helps us to fulfil our mandate.
Reaching deeper into the monetary policy toolbox
At the time of the last Watchers’ conference policy rates were already moving towards zero and we had reached a point where there was little room for manoeuvre with the standard instrument of monetary policy. We did eventually reach the effective lower bound later in the year by lowering incrementally the corridor of policy rates in June and September.
When a central bank’s ability to steer the overnight rate is limited, it can still alter its monetary stance by means of directly influencing expectations. This is the context in which the ECB introduced forward guidance in 2013. Our forward guidance was effective in flattening the money market curve and decoupling it from the curve in the United States. But its power becomes more limited, as the horizon over which the policy rate is intended to be at the effective lower bound extends beyond the forecast horizon. Moreover, forward guidance by the ECB was not and could not, in Paul Krugman’s words, “credibly promise to be irresponsible”.
What this means is that when the need for additional monetary stimulus arose, the ECB could no longer rely solely on acting in the money market, counting on transmission from the money market to other market segments, and from those to the real economy.
Like other major central banks, we therefore had to intervene directly in markets beyond the money market to have a more direct impact on the various channels of monetary policy transmission. And the way to do that was to purchase assets in those other markets.
Unconstrained monetary policy is key
Asset purchases are nothing new. They have been available and have been routinely used by central banks ever since these institutions came into existence, and that was long before the crisis started. Incidentally, asset purchases have formed part of the ECB’s toolkit from the start: outright purchases of marketable instruments – which include government bonds – have always been listed in our statute as part of our monetary policy toolbox. And the legitimacy of using public sector bond purchases in the pursuit of medium-term price stability was unanimously confirmed by the Governing Council on 22 January.
Asset purchases are unconventional, but they are not unorthodox. They are in fact eminently orthodox. And they are in a central bank’s toolbox for a reason. Independence is essential for the central bank’s credibility. But equally essential is that the central bank has the means, i.e. the policy instruments, to achieve its mandate. Together they allow the central bank to achieve price stability, which in turn boosts its credibility.
Our decision in September to make use of asset purchases had significant effects. But still, when we announced the purchase of asset-backed securities (ABSs) and covered bonds, there were some in the market place who doubted our commitment and the effectiveness of our monetary policy. They thought we might be hampered either by there being a limited availability of assets that we could purchase in the market or by legal or political obstacles to our ability to expand the range of assets, should it become necessary. If we were so constrained, that would affect our credibility because our ability to anchor expectations relies in part on the fact that we are free to set the appropriate monetary stance.
In this context, the decisions we took in January to expand the range of our asset purchases must have assuaged those concerns. We can deploy – and we are deploying – monetary policy in a way that can – and will – stabilise inflation in line with our objective.
How the expanded asset purchase programme is working its way into the real economy
As its name indicates – the expanded asset purchase programme is just an extension of the programme that we announced in September as part of a more comprehensive easing package. This package has been effective in improving the pass-through from liquidity injections into private sector borrowing costs: bank lending rates to non-financial corporations started to decline in the third quarter of last year, coinciding with the first targeted long-term refinancing operation (TLTRO) and our announcement to purchase ABSs and covered bonds, and also following the repair of banks’ balance sheets during the comprehensive assessment.
Furthermore, model-based estimates indicate that – controlling for other developments – market-based measures of inflation expectations have reacted positively to the progressive expansion of our balance sheet over the last few months. There is thus good reason to believe that as our balance sheet grows more substantially under the expanded asset purchase programme, it will support a rebound of these measures.
Second, our policy announcement was largely anticipated. On 1 January 2015, 60% of surveyed experts attached a 65% or higher probability that we would announce a public sector securities purchase programme at our January meeting. And, according to various surveys, expectations were already quite high in autumn last year.
These anticipation effects show up in the financial data. According to estimates, the impact of the asset purchase programme has accounted for most of the fall in euro area long-term sovereign yields since August last year. The same applies for movements in other financial markets metrics, such as the fall in long-term corporate bond yields of non-financial corporations.
Beyond anticipation effects, the announcement of the expanded programme of asset purchases itself also led to substantial further falls in longer-term sovereign yields. For instance, from just before our announcement on 22 January to the close of business the day after, German 20-year maturity yields fell by almost 25 basis points and Italian 20-year maturity yields fell by almost 35 basis points.
We also saw a further fall in the sovereign yields of Portugal and other formerly distressed countries – in spite of the renewed Greek crisis. This suggests that the asset purchase programme may be shielding other euro area countries from contagion, which also helps us achieve our monetary policy goals across the euro area.
The reductions in sovereign yields seem to have passed through into other fixed-income assets, as well as to equities and the exchange rate. Yields on covered bonds and corporate bonds declined in tandem with longer-term sovereign yields – even though corporate bonds are not included in our purchase programme.
Such spillover effects are associated with portfolio rebalancing, which is one of the channels through which the asset purchase programme reaches the real economy: our purchases reduce returns on safer assets. This encourages investors to shift to riskier, higher yielding assets. Pension funds, banks and other market participants that we buy securities from are likely to substitute these for other long-term assets, thereby eventually pushing up prices more broadly.
We are aware that our measures may entail some financial stability risks. But currently these risks are contained. And should they emerge, macroprudential policy is best suited to address them.
Experience with large-scale asset purchase programmes in other jurisdictions shows that the portfolio balance channel works. For instance, model-based estimates show that as a consequence of the Bank of England’s quantitative easing programme, insurance companies and pension funds invested less in gilts and more in corporate bonds,  leading to price increases of both investment grade and non-investment grade corporate bonds.
Drawing inferences from the experience in other jurisdictions is certainly helpful to gauge the potential impact of our own programme.
Much has been said about the different conditions – meaning much lower bond yields - under which we are starting our expanded asset purchase programme from those under which other central banks did so. It is claimed that this reduces the impact on bond yields. But, in fact if one standardises the size of the various programmes and takes into account anticipation effects, the overall impact of our programme on bond yields was comparable in size to that observed in other jurisdictions such as the US and the UK.
Conditions also differ because financial structures differ. In the euro area, corporate debt financing mainly takes place via banks, as opposed to capital markets in the United States. There is, however, no reason, once the transmission channel is not impaired by banks’ poor balance sheets, why this difference should impede the effectiveness of a broad-based asset purchase programme that works through a multitude of channels. The programmes of both the Bank of England and the Bank of Japan were effective and the respective economies are almost as bank-based as the euro area.
One criticism is that we should have implemented our asset purchase programme much earlier. But it is not that we have not been acting last year. In a speech in Amsterdam in April last year I laid out three contingencies that would warrant a monetary policy reaction. These were, first, an unwarranted tightening of monetary policy stance (e.g. from developments in short-term money markets) that could be tackled through more conventional measures. Second, a further impairment in the transmission of our stance, in particular via the bank lending channel, for which a targeted LTRO or an ABS purchase programme might be the right response. And third, a worsening of the medium-term outlook for inflation, which would warrant a more broad-based asset purchase programme.
As these contingencies materialised we acted, first in June with the announcement of the TLTRO and ABS purchase programme. Then, as medium to long-term inflation expectations started to drift downward in the summer and the risks of a too prolonged period of low inflation were rising, we broadened our asset purchase programme with covered bonds in September and public sector securities last January.
Summing up, market reactions both before and after our announcement, as well as experience in other jurisdictions show that the asset purchase programme can work. What is the evidence that the easing of financial conditions is finally starting to affect the real economy?
The expanded asset purchase programme and the outlook for growth and inflation
Developments are pointing in the right direction. The so-called surprise index that compares actual macroeconomic data with consensus estimates of market analysts shows that on average the latest news is positive.
The slowdown in growth has reversed. Euro area real GDP rose by 0.3% quarter on quarter in the last quarter of 2014, which is somewhat higher than previously expected. Survey evidence points to further improvements in economic activity at the beginning of this year so that the economic recovery should gradually broaden and strengthen. And in January, the euro area unemployment rate dropped to the lowest level observed since August 2012.
Of course, these improvements cannot and should not solely be attributed to our monetary easing. But our monetary policy is certainly supporting the recovery.
This is also reflected in the ECB staff macroeconomic projections that we published last week. In those projections, expectations for real GDP have been revised upwards, both for 2015 and 2016, relative to the previous exercise. These upward revisions are mainly driven by the favourable impact of lower oil prices, the weaker effective exchange rate of the euro – and the impact of our recent monetary policy measures. The latter have had a very substantial impact on what we call “market-based technical financial assumptions”, such as interest rates, exchange rates and stock prices, with the effect being especially large on long-term interest rates.
Annual HICP inflation, in turn, is expected to remain very low or negative in the months ahead and to start increasing gradually later this year. The ECB staff projections for inflation this year have been revised downwards to 0.0%. This mainly reflects the sharp drop in oil prices at the end of last year.
But there is good reason to believe that the effect of this shock will not extend beyond 2015, in part because our monetary policy decisions have significantly decreased the risk of second-round effects.
Accordingly, the inflation projection for 2016 has been revised slightly upwards to 1.5%; and for 2017 inflation is expected to be 1.8%. This expected pick-up in inflation is supported by the favourable impact of our recent monetary policy measures on aggregate demand, the impact of the lower euro exchange rate and the assumption of somewhat higher oil prices in the years ahead.
The ECB staff projections fully incorporate the estimated impact of our policy measures. They are thus conditional on the full implementation of all the announced measures. And this is indeed what we have started doing last Monday.
Let me conclude. Our recent monetary policy measures are a valid and effective tool to bring inflation closer to our policy goal. They can support a faster and more sustained recovery. This will especially be the case if they fall on fertile ground. Governments can create a more investment-friendly environment by swiftly, credibly and effectively implementing structural reforms. The beneficial impact of our asset purchases on financing conditions, rather than reducing the incentives for reforms, will actually increase the benefits of such reforms, as firms will be encouraged to increase investment, bringing forward the economic recovery. Effective, price-stability oriented monetary policy and structural reforms work hand in hand.
Cf. Michael A S Joyce, Zhuoshi Liu, Ian Tonks (2014): Institutional investor portfolio allocation, quantitative easing and the global financial crisis, Bank of England Working Paper No. 510, http://www.bankofengland.co.uk/research/Documents/workingpapers/2014/wp510.pdf