Stockholm, 18 May 2015
Ladies and gentlemen,
There is an old saying that “with money, Latin and a good horse, you can get through all Europe”. These days a good command of Latin does not help you much outside of the Vatican City. Most of us prefer to travel by plane or train. But at least in the 19 countries of the euro area, the money a citizen holds in his or her pocket crosses all national borders.
This only remains the case, however, if that money keeps its value – which is our responsibility at the ECB. So I would like this morning to talk about the measures we have taken to meet that responsibility over the last year, and to give you my view on how we should address the risks they can entail.
Responding to low inflation
From December 2011 onwards inflation in the euro area began a sustained downward drift, reaching its trough in January 2015. Responding to this situation would be challenging for monetary policy in any circumstances, but was made especially difficult because nominal interest rates were already very low.
In this context we have taken what, at face value, may appear to be a series of very unconventional measures – negative deposit rates, ABS purchases, sovereign bond purchases, to name but a few. But it is important to underscore that the objective of these measures remains thoroughly conventional: to secure medium-term price stability. And in actual fact they fit together as a coherent package that reinforces our traditional monetary policy strategy of steering inflation developments largely through the bank lending channel. This package has three parts .
The first part has been measures to increase the influence of our interest rate policy over the shape of the yield curve – especially the longer-term maturities that have the strongest link to loan pricing in the real economy.
So, as the inflation outlook has deteriorated we have both reduced our main refinancing rate to the lower bound and introduced measures to augment the impact of those very low short-term rates on longer-term rates. This has included steering downwards expectations about the expected path of short-term rates; and introducing a negative interest rate on our deposit facility, which has helped further flatten the yield curve. Overall this has measurably increased the traction of our interest rate policy on relevant market rates.
Yet it is one thing for the central bank to reduce markets rate at maturities that are pertinent to loan pricing; it is another for financial intermediaries to reflect those lower rates in the price and availability of credit for firms and households. As late as the middle of last year, viable firms were still struggling to access finance in parts of the euro area. The percentage of financially constrained but viable SMEs – defined as those with positive turnover in the last six months seeking a bank loan – was estimated to have varied from around 1% in Germany and Austria to a quarter of the SME population in Spain and as much as a third in Portugal.
The second part of our response has therefore been to improve the transmission of loose market financing conditions into the actual borrowing conditions of firms and households.
In this context, between June and September last year we launched our credit easing package to strengthen banks’ incentives to improve the availability and lower the price of credit. In parallel, we gave our full technical and operational support to the Comprehensive Assessment of bank balance sheets – the supervisory exercise aimed at forcing banks to acknowledge non-performing exposures and to raise provisions and capital where needed.
This exercise was powerful in the sense that it accelerated what had until that point been a slow process of balance sheet repair in the euro area; banks strengthened their capital by over €200 billion in advance of the outcome. In this way the new monetary policy impulse coming from our credit easing package coincided with a banking sector in a stronger position to transmit it.
But as these measures were coming into effect the euro area was hit, like all advanced economies, with a further downward shock to inflation emanating from the steep fall in global oil prices. In normal times my preference would naturally have been to look through such a development provided that it did not feed into medium-term expectations. Indeed, we had seen inflation fall into negative territory driven by oil prices in 2009, and we had not reacted because we were confident that it would be transitory. The conditions we faced in January this year, however, gave us no such comfort.
Our analysis showed that the persistence of low inflation across a range of statistical metrics was higher than in 2009. Measures of core inflation were also less sticky. But for me what was decisive was the loosening in the anchoring of inflation expectations we were witnessing, even at maturities and at horizons that we would normally expect to be more resilient to short-term inflation dynamics. This was in stark contrast to 2009 where inflation expectations hardly moved, even at the short end.
The public sector purchase programme
It was in this context that we launched the third part of our response, the ultima ratio decision to purchase assets, including public ones, as a tool of monetary policy. It was absolutely crucial that we lifted and re-anchored inflation expectations and warded off these potential second round effects. This is not only because stable inflation expectations are vital for medium-term price stability. It is also because, with interest rates at the effective lower bound, any fall in inflation expectations implies a rise in real interest rates, counteracting the credit easing we were trying to engineer.
In fact, by January 2015 expected real rates had started rising – increasing by almost half a percentage point in the previous few months alone. The cost of deflation protection had also gone up by 185 basis points between 2 December 2014 and 9 January 2015, showing that investors saw a material risk that inflation could fall further. We therefore had to react pre-emptively and decisively, and at the lower bound this leaves a single option – large-scale become the only tool to reinforce the monetary policy impulse in a way that has an immediate and meaningful impact on expectations.
While our switch to asset purchases had already begun in September with the launch of our private asset purchase programme, it was clear by January interventions in those markets alone would not be sufficient to achieve the required impact. To firmly lift and re-anchor expectations we needed to alter both the composition of our programme – to broaden the channels through which it would raise future inflation and hence affect expectations today – and the size – so that the monetary policy impulse through each of those channels would be stronger. The only markets in which we could achieve this dual effect were public sector assets.
I currently see our public sector purchase programme working through three main transmission channels.
First, it is creating strong direct price effects in markets that are key for loan dynamics. Most importantly, for loans that are priced against the sovereign yield curve nominal rates are now extremely low across nearly all euro area countries, with real rates lower still. Our purchases of non-sovereign public assets are also reinforcing a revival in credit, albeit through different channels. For example, our purchases are effectively lowering the funding costs of supranational lending institutions, such as the European Investment Bank, and of national development agencies, such as KfW in Germany. And we continue to broaden our universe of eligible agency bonds , because our aim is not to become the largest creditor on the euro area.
Second, our programme is creating portfolio rebalancing effects as investors are displaced across asset classes – affecting risk preferences – and across jurisdictions – affecting the exchange rate. This process has multiple transmission channels to real economy but one key one is through banks: on the liability side, portfolio rebalancing reduces the cost of market debt for banks, and on the asset side, it increases the opportunity cost of holding henceforth still risk-free securities, such as government bonds, over extending loans to the real economy.
Third, asset purchases have a strong signalling effect. They send a powerful signal that we will not allow price stability to be jeopardized, which helps re-anchor inflation expectations and lower real interest rates. And they also signal that liquidity will keep expanding, which supports a flattening of the term structure and further supports the easing of real interest rates and the exchange rate. The effectiveness of these signalling effects is predicated on the implementation of our programme in full, as we have communicated – that is, we will maintain the pace and volume of our intervention until we see a sustained return of inflation towards a level below, but close to, 2 % over the medium-term.
In all these ways, our asset purchase programme therefore represents a continuation and extension of our existing measures – it reinforces our credit easing and more generally the bank lending channel. In this sense, I share the view that “quantitative easing” is something of a misnomer. To be sure, the quantity dimension of large-scale asset purchases matters, but only insofar as it affects prices and hence credit conditions.
At the same time, we have to clearly distinguish instruments and objectives. In extraordinary circumstances we may need new instruments to meet our traditional objective. But what anchors trust in the central bank is that our objective and strategy stay, constant – and this is even more important when monetary policy becomes more unconventional. For this reason, innovative ideas to change our strategy, such as targeting a price level, would in fact be counterproductive in the current environment. Out goes also the misguided idea to increase our definition of price stability.
To paraphrase a well-known quote, when the facts change, we might need to change our instruments – but we do not change our mission.
Managing the side effects
In terms of reaching our objective, our package of measures is already having strong effects – perhaps stronger even than many observers anticipated. The latest data on bank lending show continued improvements in the cost of credit, the availability for credit and the demand for credit. Confidence has also notably improved, with the latest European Commission economic sentiment indicator confirming the pick-up in both consumer and business confidence. And the more confident firms and households feel in the recovery, the more credit should continue to improve.
Crucially, our intervention in January also first halted and then reversed the fall in inflation expectations. For example, the 5-year forward 5-year ahead inflation-linked swap rate has risen from its trough of below 1.5% in January to more than 1.8% now. This helped lower real interest rates by almost 50 basis points at medium-term maturities. And it has also meant that we have experienced all the upsides of oil price shock in terms of lower input prices and higher real incomes and, by preventing second round effects, none of the downsides in terms of lower wage settlements and higher real debt burdens.
Still, despite the success of our programme so far, it is important that we do not get carried away. Unconventional monetary policy must remain exceptional and time-bound. This is because the longer such measures persist, the greater the risks that can come with it. And in the euro area context there are two main risks that can in principle arise: first, greater financial instability (and let me stress here I am talking only about the situation in the euro area and not in other jurisdictions); and second, the delay of much needed structural reforms.
To provide a framework for how to think about these risks I find it useful to invoke the well-known “Tinbergen rule”, which states that a macroeconomic authority must have at least as many instruments as it has objectives.
In terms of the first risk – financial stability – one has to distinguish between conflicting and complementary objectives. In normal times, it might be appropriate for monetary policy to take financial stability into account in its assessment of the appropriate level of interest rates. The reason is that, especially if one extends the concept of the medium-term, raising rates to choke off excessive credit developments will actually contribute to our price stability objective. It helps to prevent deflationary episodes emanating from large financial crises.
And the symmetrical argument is also true: lowering rates in a downturn to support the financial sector will safeguard the transmission of monetary policy and hence support the achievement of our mandate. In this sense, I am supportive of the “leaning against the wind” argument in both directions, provided instruments and objectives are consistent.
We are in such a complementary environment today. At present we see no evidence that low interest rates are contributing to leverage-driven financial imbalances in the euro area. Indeed, the most serious financial stability risks tend to be associated with excessive developments in bank credit, and there are no signs of that in the euro area on average today.
However, were we to enter a conflicting environment, the logic of the Tinbergen rule would become more apparent.
On the one hand, if for example a bond bubble were emerging but inflation was still low, raising rates to combat that financial stability risk would run contrary to our price stability objective. On the other hand, if inflation were rising banks but banks in the euro area were still repairing their balance sheets, raising rates could have damaging consequences for the banking sector.
Given the hierarchical nature of the ECB’s mandate, if we were to face such conflicts, where price stability and financial stability point in different directions for the path of interest rates, we would have to put price stability first. This is our legal obligation. And from an economic point of view, in circumstances where financial stability effectively becomes a distinct objective from price stability, the Tinbergen rule implies there have to be instruments other than interest rates to address it.
Those instruments are of course macro-prudential policies. Following the introduction of CRDIV we now have a wide range of new tools available to the competent authorities in the euro area and, with the creation of the SSM, a stronger framework to coordinate those tools and prevent inaction bias at the national level. Should we see any indications that low interest rates are leading to financial imbalances, the first line of defence is for the competent authorities to make full use of these new but hardly tested instruments.
The division of labour is clear: monetary policy should focus on price stability; macro-prudential policy on financial stability.
Yet I am also sympathetic to the view expressed by the BIS that we need to approach macro-prudential with “a mix of ambition and humility” – that is, we need to be ambitious in using the new tools we have, but we also need to be humble in recognising their limitations. Research is so far relatively limited about the channels through which macro-prudential policies work and the magnitude of their effects. Moreover, as most macro-prudential tools are in fact micro-prudential, conflicts of competence between respective authorities could occur. This all the more that the degree of Europeanisation of micro- and macro-prudential competencies is not the same – not to speak of the competence pooling of sovereignty in monetary policy.
And in any event, macro-prudential tools tend to be more effective in building resilience against financial shocks than preventing those shocks in the first place.
For this reason, it would make sense in my view for us to remain fully committed to developing and implementing effective macro-prudential policies, but at the same time not to blindly pin all our hopes on them.
Ultimately the best defence against any conflicts between financial instability and unconventional monetary policy is for that policy to last for no longer than is strictly necessary. This means that we need to maintain the pace and volume of our interventions, as we have communicated, so that inflation rises back towards 2% as quickly as possible and that monetary policy can begin once more to normalise.
But it also means something more: if interest rates are low because the natural rate is low, then for interest rates to rise back to more normal levels the natural rate will have to rise as well. And insofar as a low natural rate reflects weak investment demand caused by diminishing labour supply and low productivity growth, this may only be possible in a context of structural reforms aimed at boosting supply capacity. This brings me to the second risk of unconventional monetary policy: that it could in principle delays the very reforms that are necessary for monetary policy normalisation.
Let me say upfront that the evidence we can see in the euro area at the moment suggests that this is not necessarily the case. Some countries have recently taken steps in the right direction – but more can and needs to be done.
My hope is that governments are recognising that accommodative monetary policy and low oil prices, rather than removing the need for reforms, creates the perfect environment to do them. Research suggests that reforms can be more costly in a weak macroeconomic environment and, in particular, when there are serious credit constraints. With our successful credit easing policy we are progressively removing each of those barriers.
More generally, however, I would question how much monetary policy decisions should be based around reform incentives. Applying again the Tinbergen rule, this creates two objectives to be attained with one instrument – and just as for financial stability, in the current environment those objectives conflict. The ECB has to prioritise price stability rather than acting as the censor of governments. Euro area governments as a community must ensure that each economy meets the requirements of membership. And it is in this context that I and other ECB Executive Board members have recently called for a stronger Economic Union in the euro area with improved common governance of structural reforms.
Let me conclude.
The diminishing price pressures over recent years have required the ECB to act forcefully and repeatedly to fulfil its mandate. This culminated in January 2015 with our decision to expand our asset purchase programme to stave off deflationary risks and stop the fall of inflation expectations. In doing so we have sent a strong signal that we will safeguard price stability no matter what. This is our mandate as enshrined in the Treaty. And although our instruments have changed our conviction and mission have not.
Some might wish that we use our armoury for other purposes. But we will not be moved. We are an independent institution. Our rules are non-negotiable and we do not act for the benefit of interest groups or individual countries. Nor can financial stability considerations override our primary objective. Monetary dominance will always prevail.
Compliments of European Central Bank Directorate General Communications