A consistent strategy for a sustained recovery

Lecture by Mario Draghi, President of the ECB,
at Sciences Po,
Paris, 25 March 2014

The euro area has been through a crisis of almost unprecedented drama and severity. Like the Great Depression of the 1930s, this will most likely provide rich material for economic historians for decades to come. It will be studied and analysed, its causes and consequences debated and revised. So it is interesting for us to imagine, when the dust has settled, where the various accounts of the euro area debt crisis will locate its turning point.

Some historians may locate it in individual policy measures taken during the crisis; others may place it in a future that we have not yet reached. In my view, the turning point has passed, and it came in the summer of 2012. What changed at that point was that crisis management shifted towards the development and execution of a consistent recovery strategy. It is this strategy that I am going to outline in my remarks today.

I will organise my remarks as follows. First, I will describe the initial development of the crisis, illustrating how policy choices made under the pressure of events and that were commendable by themselves, but that were sequenced in the wrong order, made dealing with the consequences of the debt overhang more difficult. This interacted with features of the euro area’s institutional structure to postpone the recovery.

Thereafter, I will describe how the right sequence of steps after June 2012, when the banking union project was first agreed, has put the euro area back on a trajectory towards recovery. The first step was ‘rebooting’ the financial system, which is a necessary condition of a sustained recovery, not least because it helps monetary policy to manage aggregate demand. But it is not a sufficient condition: policies of structural reform that lift the level of potential growth are an equally important part of the recovery strategy.

In describing this strategy, I am not only talking about the past, but also about the present and the future. The crisis is not over. To be successful, the recovery strategy is being, and must continue to be, executed with commitment and perseverance.

Looking back at the crisis

The early phase of the crisis followed roughly the same pattern across advanced economies. Most had been through a long period of excessive debt accumulation, with private debt levels in several jurisdictions reaching historical highs. In some countries this fuelled over-investment in unproductive sectors, including real estate, which is the sector associated with most of the severe financial crises in history. The mirror image of this development was a credit bubble in the banking sector, as banks financed both housing supply, i.e. loans to real estate developers, and housing demand – that is, mortgages.

When the crisis broke out, this debt build-up quickly became a debt overhang. Many advanced economies entered a prolonged period of deleveraging as firms, households and banks attempted to reduce their debt levels. For firms, this meant less investment. For households, less consumption. And for banks, less credit. Moreover, the collapse of Lehman Brothers sent shockwaves through the global financial system leading to an unprecedented rise in uncertainty and risk aversion. This led to a further retrenchment in investment, trade and hiring and a globalised “Great Recession”. Government budgets went deep into deficit to offset this massive shock to nominal spending.

These events were not only shared across major economies, but they were also largely consistent with historical experience. There have been several episodes in economic history of large build-ups followed by periods of debt deleveraging. Serious financial crises tend to be followed by slow economic recoveries.

Nevertheless, by mid-2010, most advanced economies were showing signs of returning to growth, albeit at a slow pace. At this point, however, the trajectory of the euro area departed from others. While the recovery gained ground in the US in particular, the euro area entered into a second recession that lasted until the second quarter of 2013. Why did this divergence happen?

For two reasons that were specific to the euro area. First, the sequencing of policy responses after the first bail-out for Greece aggravated concerns about bank and sovereign debt sustainability. Second, these concerns interacted with an incomplete institutional framework in a self-reinforcing way.

In addressing the situation in mid-2010 – and with the benefit of hindsight – one could have legitimately expected the following sequence of actions. First, agree on a solid backstop for dealing with sovereign and banking sector problems. Thereafter, conduct a stress test and recapitalise banks where necessary. Then, with banks in a stronger position to absorb losses and a sovereign backstop in place, construct a consistent framework for dealing with sovereigns with excessive debt. Finally, apply that framework to countries that were deemed to need it. For example, this sequence from backstop to stress test to recapitalisation was applied in the US, once the lessons from the Lehman Brothers shock had been drawn, and it accelerated the clean-up of the country’s banking system.

However, in 2010 and 2011 it was almost the reverse sequence that took place in the euro area. The Deauville agreement on private sector involvement in October 2010 and the Greek debt restructuring in July 2011 were announced while an effective backstop for solvent governments was still being constructed. And the initial stress testing of banks in 2011 and the capital raising exercise in October that year were conducted without any clear backstop for solvent banks. The effect was to cause many banks and some governments to come uncomfortably close to losing market access, or to lose it altogether. As a consequence, instead of acting as a shock absorber, both banks and governments began to act pro-cyclically.

This situation then interacted negatively with two features of the euro area’s institutional structure.

The first was the euro area’s incomplete financial integration. While prices had converged in many asset classes prior to the crisis, it turned out that the euro area had not in fact created the conditions for deep cross-country financial integration. Integration was largely based on short-term interbank debt, rather than on equity or direct cross-border lending to firms and households, and under stress it quickly unravelled. Indeed, the build-up of financial imbalances in the euro area periphery had in part been financed by short-term lending from banks in the core. Those banks then quickly reversed their exposures in what amounted to a “sudden stop”, analogous to crises experienced by emerging economies. This contributed to the fragmentation of the euro area banking sector and economy along national lines, a phenomenon which had not been visible in the early stages of the crisis.

The second feature was the euro area's fiscal framework, which was not strictly enforced. That framework was explicitly designed to ensure fiscal responsibility, with two resulting benefits. First, if strictly applied, it would create fiscal space ex ante for governments to absorb exceptional shocks, such as the one experienced worldwide. And second, by anchoring confidence in the medium-term soundness of public finances, it would allow governments to run those counter-cyclical policies while retaining market access. In such a situation, there is no need for fiscal risk-sharing or a fiscal backstop.

Yet, as the fiscal framework was not strictly enforced, the two benefits were reversed: a number of governments either did not have the fiscal space to absorb the shock they faced in the early stage of the crisis (consider Belgium or Italy); or they were unable to maintain the trust of the market while doing so (consider Portugal or Ireland). Fiscal policy in these countries therefore had to switch from providing a counter-cyclical buffer to convincing investors of debt sustainability.

Under these circumstances, it was unavoidable that fiscal consolidation was front-loaded. But it also meant that fiscal policy changed from being a tailwind to a headwind, and added to the drag coming from deleveraging across the private sector. In particular, as the banking sector had not yet been cleaned up and strengthened, lower growth produced a further deterioration in balance sheets and added to pro-cyclicality.

In managing the crisis, European policymakers faced an unprecedented set of circumstances. They were making decisions in real time in the face of political and institutional constraints. So the aim of my comments is not to criticise. Rather, it is to highlight the fact that the sequencing and consistency of policy decisions matter. Policymakers dealt with the immediate situation without simultaneously addressing all its consequences. It was only when this began to change in June 2012 that we returned to the path of recovery.

The beginnings of a consistent strategy

What happened at this time was that European policymakers acknowledged the need to complete the euro area’s institutional architecture, the initial stage of which was setting up the banking union. And in doing so, they initiated what I believe was the necessary first step of a consistent strategy for a sustained recovery.

The banking union had to be the first step of a longer sequence, for two reasons. First, because it was necessary to consolidate the single currency. Second, because it provided an opportunity to “reboot” the euro area banking system, which in turn is a pre-condition for the recovery.

Let us consider the two in sequence.

Restoring the singleness of money

Money, it has to be remembered, is a liability of the banking system. Banknotes represent only a fraction of the money we use daily. The bulk of money is deposits, which are a liability of commercial banks. So for there to be a truly single money among sovereign countries, there has to be fungibility of deposits across borders.

Yet, this fungibility came under threat during the crisis. It was threatened initially by the fragmentation of financial markets in the euro area, and then exacerbated by the emergence of redenomination risk in financial prices. Those unfounded fears of redenomination put price stability at risk, which the ECB had to alleviate through the creation of its Outright Monetary Transactions (OMT) programme. Our actions underlined the irreversibility of the single currency and were decisive in restoring confidence.

But the underlying drivers of fragmentation still remained, especially the emergence of credit risk premia at the national level that mirrored the perceived credit risk of sovereigns, in particular after the Greek debt restructuring. This link between sovereign and bank risk reflected, first and foremost, the perception that the ultimate guarantor of deposits in the banking system is the state. Hence, where perceptions of sovereign creditworthiness diverged, so did confidence in their respective banking systems.

In essence, what these developments were highlighting was that we did not have a truly single banking system in the euro area; we had a juxtapositionof national banking systems, which is why they fragmented so easily. To ensure that the single currency truly is single, therefore, the only realistic option was to bring together those national systems into one single system, so that the fungibility of deposits was re-established. This is where the banking union comes in.

Banking union means three things: it means a single supervisory framework that minimises equally the risk that a euro area bank takes excessive risk and runs into failure. It means a single resolution framework, so that if a bank does still fail, it can be resolved in the same way, with limited use of taxpayer money, irrespective of where the bank is located or the fiscal strength of its government. And it means a system of deposit protection that provides depositors with equal confidence that their deposits are safe, regardless of jurisdiction.

We are now well advanced in the process of unifying the banking system in this way. The Single Supervisory Mechanism (SSM) will begin operating in November. A deal was reached last week on a Single Resolution Mechanism and Single Resolution Fund. And a harmonised approach to the level and funding of deposit guarantee schemes across the euro area has been agreed, as a first step towards a single deposit guarantee scheme. Deposits of individuals and small- and medium-sized enterprises (SMEs) will also have seniority in any future bank resolutions. Together, this goes a long way towards creating a genuine banking union, although some important details, such as the backstop for European resolution financing, still need to be clarified.

By reinforcing the singleness of money, the banking union provides the conditions for a lasting reintegration of the single financial market. It is therefore a pre-requisite for the recovery. It does not, however, by itself generate a recovery, nor does it put banks in a position to properly support that recovery. This brings me to the second step in the sequence of events required to achieve that goal: the role of the SSM in cleaning up the banking system.

Cleaning up the banking sector

Prior to the crisis, euro area banks had entered a rapid period of balance sheet expansion. From the start of that expansion in 2005 to its peak in 2012, banks assets increased by more than 60 percentage points of GDP. This was associated with the development of unsustainable bank business models. Banks relied too much on debt to finance their lending, and that debt depended too much on wholesale market funding and too little on deposits.

This model was only able to develop because of the perception of an implicit state guarantee for bank debt – a perception that reinforced the link between sovereign and bank risks that I described above. The deterioration of sovereign credit, on the one hand, and the clarification of the rules regarding bail-in of bank debt, on the other, have both helped to bring to an end a funding model that was neither desirable nor sustainable.

The euro area banking system is therefore now undergoing a process of restructuring and deleveraging. As this is a necessary correction, it is not a process that policymakers should seek to prevent. However, it is a process that needs to be properly managed.

Deleveraging can essentially take two forms: a “good” form and a “bad” form. The “good” type is where banks quickly carve out non-performing or non-core assets and raise equity, allowing them to restart lending to new, creditworthy clients. The “bad” type is where they sell good assets and hold on to non-performing assets in the hope that their value recovers. This tends to create so-called “zombie banks” and leads to a prolonged period of low credit growth. And this can be even more damaging if, due to fears about counterparty credit and liquidity hoarding, liquidity dries up and banks have to sell assets at distressed prices. Such an interaction of market risk and funding risk can cause a systemic crisis, with an outright credit crunch as the consequence.

Historical precedents suggest that – all other things being equal – a quick, “good” deleveraging tends to bring about an earlier recovery.

In the euro area we have largely avoided the worst form of deleveraging, thanks to ECB interventions to provide liquidity to banks at key moments of the crisis, notably our move to unlimited liquidity provision in 2008 and our two longer-term refinancing operations (LTROs) in late 2011 and early 2012. In this environment banks have made progress in deleveraging and restructuring. However, as late as last year there was still uncertainty as to the true extent and quality of this deleveraging. This was shown by persistent investor doubts about bank asset valuations, and low credit growth for the real economy.

It was in this context that the creation of the SSM became critical. The SSM provided a catalyst for the clean-up of the banking system to take place more quickly and more comprehensively than would otherwise have been the case.

Why was this? Introducing a joint supervisory framework requires us to conduct a thorough health check of bank balance sheets, using harmonised methods, so that any legacy issues are recognised, and that all banks enter the new supervisory regime with sound levels of capital and liquidity. We are currently conducting this comprehensive assessment in cooperation with national supervisors. To illustrate its scale, the total amount of risk-weighted assets (RWA) being reviewed amounts to €3.7 trillion, or just below 60% of total RWA for the 128 participating banks.

The nature of the assessment, which includes an asset quality review, is conducive to achieving the “good” type of deleveraging. It will shed light on the composition and valuation of bank assets, which ought in turn to trigger fresh equity raising where appropriate, and a carving-out of assets where necessary, either by selling to the market or via “bad banks”. Raising equity and strengthening balance sheets is in the long-term interests of bank shareholders, even if it implies an initial dilution of their claims.

In fact, the cleaning-up of bank balance sheets does not need to wait for the end of this assessment. The best outcome is one where banks are forward-looking and take any corrective action before the end of the process. And indeed, we have seen in many cases that banks have revised asset valuations, raised equity and disposed of assets. Bank balance sheets declined by around 20 percentage points of GDP in 2013 alone, partly in anticipation of the assessment. Increasing robustness has been mirrored in banks’ share prices, which increased by around 40% in 2013 relative to average market growth of 20%.

Over time, I expect this process, together with the new regulatory framework which is now being completed by the Basel Committee, to result in a euro area banking system with a more streamlined balance sheet structure and higher levels of capital. We should be wary of situations where bank assets are many multiples of GDP, as we saw in countries like Ireland and Cyprus before the crisis. Bail-in ability will likely also mean that senior unsecured bank debt plays a lesser role in the future relative to deposits. In other words, this will be a banking system that comes closer to its traditional role of taking deposits and making loans.

However, a permanent shrinking of the banking system raises some questions about how finance will be intermediated in the euro area. If intermediation between savings and investment is taking place less through banks, then it must take place elsewhere, through capital markets.

Capital markets are already increasingly complementing banks in the euro area: as bank credit to corporates has fallen, it has been approximately matched by issuance of corporate bonds. However, the substitution is not perfect, in particular for smaller firms, for which it is more costly for investors to obtain adequate information. This means that, absent policy action, SMEs are at risk of losing some access to finance.

If we are to continue with a consistent set of steps to support the recovery, cleaning up the banking system therefore has to be accompanied by policies to facilitate the development of capital markets, especially for smaller firms. It is for this reason that we have frequently highlighted the importance of creating conditions that support capital market deepening, including by removing regulatory obstacles to the least risky models of securitisation.

Moving towards a sustained recovery

This “rebooting” of the financial system, together with the construction of the banking union, had to be the first step in a consistent strategy, as a healthy banking system is a necessity for a sustained recovery. Yet it is still only an enabling factor. What it achieves, essentially, is to create an environment where “normal” macroeconomic management can resume. It is this that generates the recovery.

The banking sector supports this process through two channels. First, it facilitates the ability of monetary policy to maintain price stability and help economic growth return to its potential. Second, it ensures an adequate supply of financing for investment, and an efficient allocation of credit between productive and non-productive firms, thus also supporting potential growth.

Let me explain these two points in more detail.

Maintaining price stability and closing the output gap

The task of stabilising output around its potential to maintain price stability falls within the responsibility of the central bank. This has been especially true in the current crisis as, due to the unavoidable consolidation of public finances that I described earlier, the contribution of fiscal policy to economic stabilisation could not be as significant as in normal circumstances. A larger burden was therefore placed on monetary policy to achieve price stability by managing aggregate demand.

To illustrate this, we have not only cut interest rates to our lowest level since the creation of the euro, but we have also introduced forward guidance, whereby we commit to keep our policy interest rates as low as they are currently or even lower for an extended period of time. This supports the economy today by steering expectations about the future. In particular, our forward guidance implies that short-term real rates, which are negative today, will become even more negative in the foreseeable future. This is because our policy rates will remain low or lower in nominal terms, while inflation is projected to gradually pick up. In turn, expectations of lower short-term real rates are reflected in the level of medium-term real rates, those that are most relevant for investment decisions by entrepreneurs. In this way, our guidance on the level of rates tomorrow, and the day after tomorrow, supports investment today.

However, these stimulatory effects have been hindered by impairments in the transmission of our monetary policy to all parts of the euro area during the crisis, in particular as regards the bank lending channel. The responsiveness of bank lending rates to cuts in our main interest rates has been uneven across jurisdictions, with lending rates in some countries falling more or less in tandem, and in others reacting barely at all. The main explanation for this is the fragmentation of banking systems and bank deleveraging I described above.

With funds no longer flowing freely within the euro area, banks had to pay higher rates to attract deposits, pushing up their funding costs. Higher funding costs meant higher lending rates. Banks in the process of rebuilding their capital also required greater compensation for loans that carried higher risk weights, such as those to non-financial corporations. And they needed to rebuild profit margins to cope with a high-risk environment. This again implied that low ECB interest rates could not be fully passed on to borrowers.

Yet this also implies that as the effects of our OMT announcement feed through, and as policies to reverse fragmentation accelerate, and bank deleveraging and restructuring proceeds, monetary policy should become increasingly effective.

In fact, we are already seeing evidence of that process. Deposit rates in the euro area have been converging towards our policy rate. And banks that are more advanced in their balance sheet adjustment have been better able to reflect our policy impulse in their lending volumes and rates. If bank lending rates in the euro area periphery ultimately converge with those in the core, this would effectively represent a significant additional monetary easing.

As such, I expect monetary policy to regain influence over the economic cycle, and our accommodative stance to support a gradual closing of the output gap in the coming years. This is reflected in the current ECB staff projections, which foresee inflation rising to 1.0% in 2014, 1.3% in 2015 and 1.5% in 2016. If any downside risks to this scenario appear, we stand ready to take additional monetary policy measures that ensure our mandate is fulfilled. In other words, we will do what is needed to maintain price stability.

Our accommodative monetary policy has an ancillary advantage in the current situation: it facilitates the ongoing process of deleveraging. It both reduces the numerator of debt ratios, i.e. the nominal interest rate, and raises the denominator – that is, the nominal growth rate. And this is again facilitated by the reduction in financial fragmentation, which tends to increase nominal rates while decreasing growth rates.

Raising the potential

Returning output to its potential is one part of the recovery story; but equally important is the level and trend of potential itself. In the initial stages of the crisis, the euro area, like most advanced economies, received a substantial downward shock to potential growth. While measuring potential growth is complicated, the projections of all the major economic institutions put the euro area’s rate of potential growth lower than before the crisis.

Low potential growth creates several problems. It implies stagnant real incomes, thus limiting welfare gains and complicating the management of the debt overhang for firms and households. It is accompanied by high levels of structural unemployment, which is both socially very costly and creates a drag on aggregate demand. And it reduces the amount of stimulus that monetary policy can provide to the economy without generating inflation.

For all these reasons, engineering a boost to potential output is the key policy challenge facing the euro area today – and it is an essential part of a consistent recovery strategy.

However, raising potential growth is not a task for monetary policy. Monetary policy can act on the output gap, but it is structural policies that act on potential output. And given the euro area’s weak demographics, this has to be principally achieved by policies that raise productivity, which in turn comes down to raising investment. Investment is crucial because it is today’s demand and tomorrow’s supply. Yet, investment in the euro area has fallen by 18% during the crisis, and 35% in stressed countries.

This is where the process of repairing the banking sector can help lift potential growth as well. Evidence from euro area firms shows that financial constraints have strongly affected firms’ investment decisions during the crisis. This has come both from the supply side – banks that will not lend to firms – and the demand side, i.e. firms that do not want to borrow because their balance sheets are too leveraged.

With better capitalised banks, we should see a steady resolution of both these issues. Financing should resume for firms with profitable investment opportunities, which should benefit in particular young firms that create most net jobs and are most reactive to changes in investment opportunities. At the same time, acknowledging losses and raising capital is a pre-requisite for banks to restructure loans to distressed borrowers, thus facilitating a gradual workout of private sector debt.

In turn, this will allow the banking sector to support the Schumpeterian process of creative destruction that is key to productivity growth – that is, allocating and reallocating resources towards those firms and sectors that most generate innovation. Recent micro-level research by the Eurosystem’s CompNet network has revealed that the distribution between the most and least productive firms in euro area countries is very large, which implies that reallocation could provide a significant boost to productivity.

Completing the picture

Yet this is not the end of the story. The rebooting of the financial system is necessary but not sufficient to achieve higher investment. Investment is also low because demand is low and there is slack in the economy. This will partially unwind as the economy recovers, but we cannot exclude that investment will not return to its former level as certain sectors have permanently downsized as a result of the crisis.

This raises the question of whether policy is doing enough to compensate for these structural changes – in particular, whether it is creating the framework conditions for new business models to emerge that require higher levels of productive investment.

For example, is regulation in all countries conducive to investment? 45% of fixed investment in Europe is estimated to be concentrated in sectors where governments have significant regulatory influence, implying that regulation could play an important role in boosting investment demand. Even at the firm level, regulation can influence whether the most productive firms are allowed to grow. One well-known distortion in this regard is the change in status and requirements for firms in this country when they have 50 employees or more.

Is the tax code conducive to investment? What matters for investment demand is the after-tax return that firms receive, meaning tax incentives can improve the risk-return profile of investment projects, and even more so in a lower growth environment. In this country, several official reports have suggested that high corporate and labour tax rates hinder investment. The focus of the current government on this issue is therefore fully justified.

Is the education system conducive to investment? Europe’s comparative advantage, like all advanced economies, is increasingly located at the high-end of the value chain. Yet investment can only flow there if the workforce has the skills to match. Around half the countries in the euro area rank below the OECD average for student performance in mathematics, reading and science, implying standards could still rise. Indeed, educational attainment is no longer only a factor in income, but also in employment prospects: at the end of 2012 18% of workers with low education levels were unemployed, compared with only 6% of highly educated workers.

All this underscores that, when we think about how to restore competitiveness in the euro area, we should not only think in terms of lowering wages relative to productivity. This may be a necessary short-run measure in some jurisdictions, but it is a zero-sum game: some countries benefit at the expense of others. The ultimate source of competitiveness for the euro area has to be productivity gains built on investment, and this is a positive-sum game. It raises welfare for the euro area on aggregate.

What I am saying here is not new. The need for a continuous rise in productivity performance has long been recognised in Europe; it was the context for the launch in 2000 of the Lisbon Agenda, which aimed to make Europe “the most competitive and dynamic knowledge-based economy in the world” by 2010. As we know, most of its targets were not reached. But the objective was, and remains, fundamentally the right one.

What is different today, compared with a decade ago, is that weak productivity can no longer be hidden by increasing private leverage or rising public sector debt. There are no levers that can be pulled to raise real growth other than those that address the structural barriers in our economies. Fiscal policy can support growth by ensuring that consolidation has a better composition – less focused on raising taxes, more on prioritising expenditure – but it makes no sense to unwind consolidation now that the hard work has been done. This is why I reiterate that there is no bigger policy challenge for the euro area than raising potential growth.


Let me conclude.

The point I have tried to emphasise today is simple: to navigate through a debt crisis such as the euro area has experienced, the sequencing and above all the consistency of policy choices is everything.

We need to be consistent across space, and we need to be consistent across time. There are indeed two inescapable lessons from the account of events that I have presented to you today.

First, a full recovery will be achieved when – and only when – we complete in full the sequence of steps I have highlighted. With each step in this sequence, we should not ask “Have we done enough?”. It will be clear when we have done enough: when debts are reduced, when potential output has been raised and when unemployment has come down – while all along preserving the integrity of money, meaning price stability.

This means, ultimately, looking hard at whether our economies are fit for purpose in a specialised, globalised knowledge economy. This is something that national governments, businesses and social partners have to do together. What is being done here in France is important in this respect, as a strong euro area needs a strong French economy.

The second lesson is that the recovery stems from joint action. We have seen the effects of the fragmentation of financial markets across national lines. We have experienced the consequences of confidence diverging across countries. We have suffered from the resulting loss of growth and rise in unemployment. No one who has experienced either this crisis, or the economic volatility of the 1970s and early 1980s, can credibly conclude that a return to a mere juxtaposition of national policies will improve the situation of any of our countries.

Rather, it is by making our policies consistent across borders that we have achieved positive results, and it is by heeding this lesson that we will continue to do so. In some areas, such as the banking union, this means formulating policy centrally. In other areas, such as fiscal policy or structural reform, it means accepting effective scrutiny from peers. In both cases, this is not a loss of sovereignty. In fact, I regard it as a recovery of sovereignty, because it is the means to provide once again, and sustainably, the stability and the opportunities that citizens demand from policymakers.

Winston Churchill said that “to achieve great things, two things are needed: a plan, and not quite enough time”. I hope that I have made clear today that we do have a plan. And since we certainly have no time to spare, I trust that, if we remain resolute, great things for the euro area and its citizens can become possible.

Speaking engagements

Media contacts