* * *
Deleveraging is not a process that policy-makers should seek to avoid; rather, it needs to take place and be properly managed, explains ECB Executive Board member Benoît Cœuré. Describing the good, the bad and the ugly deleveraging at the Future of Banking Summit organised by “The Economist” in Paris, he emphasises how the wrong type of deleveraging could impede recovery and make it more difficult for central banks to engineer an appropriate degree of monetary accommodation.
To ensure the right type of deleveraging is achieved, Mr. Cœuré highlights, “both monetary policy and prudential supervision – notably the ECB’s ongoing comprehensive assessment – have a key role to play.” According to him, the comprehensive assessment of banks’ balance sheets and the steps taken towards the banking union are crucial to establish the conditions necessary for a transparent, competitive and stable banking sector. He adds, “and they will help monetary policy to regain traction across the euro area.”
He concludes by explaining why it is important for the banking union to be complete, “Not moving promptly towards a single resolution mechanism (SRM) and a single resolution fund (SRF) would prolong financial fragmentation, leave the financial system exposed to systemic fragility, and, ultimately, be harmful for growth and jobs. This would in particular be the case if the SRF is only slowly mutualised, and if it cannot resort to a common European backstop from the outset.”
* * *
Ladies and Gentlemen ,
It is a great pleasure to speak to you at the Economist’s “Future of Banking” Summit.
But before turning to the future of banking, let me briefly look back at its recent past. As you know well, the euro area banking sector is currently undergoing a secular deleveraging process. The need for deleveraging had its roots in a period of strong bank balance sheet expansion in the last few years.
From the start of that expansion in 2005 to its peak in 2012, the total assets of monetary financial institutions rose by more than 60 percentage points of GDP to €33.7 trillion, i.e. 3.55 times euro area GDP. But as the crisis highlighted, this expansion was not necessarily backed by rising productive capacity in the euro area economy. Instead, it was, at least partly, channelled to low-productivity activity, as manifest for instance in the overinvestment in the construction sector in some Member States.
The current adjustment reflects, therefore, the aftermath of a major financial crisis and the debt overhang that inevitably comes with it. It also reflects the end of the unsustainable business models and liability structures that many euro area banks had developed before the crisis. Hence, deleveraging is not a process that policy-makers should seek to avoid; rather, it needs to take place and be properly managed.
The main point I want to make today is that how this process is managed has important implications for monetary policy. Deleveraging badly hinders the pass-through of our interest rates to borrowers. But by the same token, deleveraging well, to the extent that it restores the transmission of monetary policy, could potentially unleash a pent-up monetary easing. Achieving the right type of deleveraging depends crucially on the incentives embedded in the institutional framework in which banks operate. Both monetary policy and prudential supervision – notably the ECB’s ongoing comprehensive assessment – have a key role to play in this respect.
Three types of deleveraging and their effects
Broadly speaking, there are three different types of bank deleveraging. These are the good, the bad and the ugly, as The Economist itself stated in a blog a couple of years ago. 
“Good deleveraging” involves banks raising capital and getting rid of impaired assets in a targeted and rapid manner. This approach recognises and corrects the flaws in legacy business models and allows for a relatively swift reallocation of credit to more competitive sectors. This eventually translates into a more dynamic recovery.
“Bad deleveraging” entails an indiscriminate reduction in balance sheet size, regardless of the quality of assets. The main motivation for such adjustments is to reduce indebtedness. And in order to “balance the books”, this process is accompanied by asset shedding. The upshot is a potentially prolonged, across-the-board reduction of credit to the real economy.
Finally, “ugly deleveraging” comprises banks specifically discarding good assets while keeping distressed assets in the banking book and recording them at close to nominal values. This method can rapidly turn a bank into a “bad bank” that does little to support credit creation.
These last two types of deleveraging can have significant real economy effects. Financially weak banks tend to roll over unprofitable loans that do not support growth – which is why they have sometimes been dubbed “zombie banks”.  There is in fact some evidence that such evergreening practices have appeared in the euro area as well in the last few years. 
The wrong type of deleveraging also has a further effect, which I want to concentrate on today: it harms the transmission of monetary policy.
Given the bank-based nature of our financial system, the main channel through which the ECB’s monetary policy impulse reaches the real economy is through bank lending rates. And this is particularly important for smaller borrowers, as they cannot easily substitute bank financing for capital market financing such as corporate bond issuance, as larger borrowers have been able to.
But banks that are deleveraging through a prolonged process of retaining earnings and rolling-over bad loans tend to block this channel. Regardless of the level of nominal interest rates, they are likely to demand higher compensation for new exposures that increase risk-weighted assets, including notably loans to non-financial firms. And if they are competing for new deposits to improve their loan-to-deposit ratios, they will face higher funding costs, which then have to be incorporated into the interest rate spread for new loans. All this means that monetary policy starts to lose traction.
Indeed, different deleveraging needs are one explanation for the fragmentation in the pass-through of interest rates in euro area countries. Those countries that saw the highest pre-crisis increases in private debt – principally Ireland, Cyprus, Portugal and Spain – and that therefore have the greatest bank deleveraging needs, are by and large the same countries where monetary policy transmission has been most impaired. Of course, other factors such as sovereign risk have also played a role.
Significant progress has already been made with bank deleveraging. Banks have engaged in a sweeping shrinkage of balance sheets – declining by around 20 percentage points of GDP in 2013. Loan-to-deposit ratios have also come down, from more than 140% in 2008 to less than 120% at the end of last year, mainly through higher deposits. And we have observed a marked increase in banks’ risk-weighted capital ratios since the beginning of 2009. Since the onset of the global financial crisis, euro area banks have issued around €235 billion of equity, in addition to other forms of capital (e.g. retained earnings, CoCos, etc.). As a result, large parts of the sector now comply with the minimum capital requirements under Basel III – well before it becomes binding.
There are already signs, such as the stabilisation of credit conditions , that these efforts are improving the transmission of our easy monetary policy stance. The challenge for policy-makers now is to finish the job: to ensure that deleveraging takes place fast, that it takes place well, and that it happens everywhere where it is needed.
Monetary policy intervention
Monetary policy itself has a role to play in this process. Our measures have helped ease pressures on the liability side of bank balance sheets that might have led to “bad” or “ugly” deleveraging. For example, in the first phase of the crisis after the Lehman Brothers failure, banks faced a funding crunch which could have led them to fire sales of assets. We helped avert this by effectively standing in for the interbank market. Our two very long term refinancing operations (VLTROs) in 2011 and 2012 had an analogous effect in the term funding market.
It is important to stress, however, that we aimed to prevent a disorderly deleveraging, not the deleveraging process itself. Bank deleveraging leads to a reduction in inside money, that is money created by banks, and with fragmented financial markets, liquidity flows within the euro area were disrupted. By providing an elastic supply of central bank money, we substituted inside for outside money to support the stabilisation of the banking system. Indeed, base money – i.e. the total amount of currency in circulation and banks’ holdings of liquid deposits with the Eurosystem – increased by almost 30% following the two VLTROs.
The ECB’s forward guidance should also be supportive: the Governing Council of the ECB expects interest rates to stay at present or lower levels for an extended period of time, while inflation should rise towards 2% over our projection horizon. This implies that real interest rates for borrowers will progressively fall as inflation rises.
And we stand ready to act if this scenario does not materialise.
But, nonetheless, broad money and credit expansion by the euro area banking system has remained anaemic. The annual growth rate of M3 slowed steadily in the aftermath of the VLTROs and stood at a very low level of 1.2% in January 2014. The monetary policy impulse, as reflected by the increase of base money, has hardly been transmitted to the real economy.
Additionally, it is clear that the ingredients of “good deleveraging” cannot be provided by monetary policy alone. Good deleveraging involves, on the liability side, equity being built up through outright issuance or as last resort through public interventions. And on the asset side, it involves writing down and carving out non-performing assets. Indeed, without such measures there is a risk that low interest rates and ample liquidity provided by the ECB might delay the restructuring and reallocation that are necessary to escape a debt overhang. For this reason, prudential measures must also play a key part in the deleveraging process.
Prudential policy in the new banking union framework
For the euro area, the principal tool to steer deleveraging towards its good form is the ongoing comprehensive assessment of banks’ balance sheets. The ECB has recently released more technical details about an important component of this exercise, the asset quality review, so I will not repeat its features today. The assessment will shed light on bank assets and ensure that non-performing exposures are fully recognised, and prompt timely corrective action in the form of bank restructuring and capital replenishment.
In turn, I expect this to help monetary policy regain traction across the euro area. Actually, this for me will be a measure of the success of the exercise.
Why do I expect this?
For banks that are found to be healthy by the assessment, the price mechanism will no longer be obscured by uncertainty about asset valuation and bank funding models. This should be reflected in higher share prices and cheaper funding costs, supporting the pass-through of our interest rates.
For banks that are restructured and recapitalised, lending rates should also become more responsive to our monetary policy impulses. There will no longer be a need for such banks to rebuild capital, and hence their incentive to reduce loans that carry higher risk weights, like those to non-financial corporations, will disappear. When loans are repaid, banks can opt to lend again rather than to shrink their balance sheets.
At the same time, spreads on loans should come down. Funding pressures are likely to attenuate as banks reach their target loan-to-deposit ratios and as interest rates on deposits converge. And as more banks free up balance sheet space for new lending, competition among banks for borrowers should start to increase, putting further downward pressure on new lending rates. We have already seen some of these dynamics playing out in Spain since mid-2012.
By identifying weaker and stronger banks, and putting a price on bank assets, the comprehensive assessment may also have a further effect: the results could present an opportunity for a revival of M&A activity, which has been very weak since the crisis, with the overall value of deals decreasing fourfold from 2008 to 2012. A rationalisation of the euro area banking sector will increase efficiency, and hence be conducive to a smoother transmission of monetary policy.
In short, a strict and credible comprehensive assessment will significantly support monetary policy. Insofar as it leads to bank lending rates converging between the core and the periphery, it will allow a significant additional easing of monetary policy in some jurisdictions. And insofar as it leads to better access to finance, higher growth and rising inflation, it will support our expectation that real interest rates will gradually ease over the projection horizon.
One final point: looking further ahead, I expect the downsizing of the banking sector to accelerate the development of alternative, capital market-based sources of finance, especially for smaller firms. We are already seeing this playing out in some jurisdictions – for example, the “mini bonds” scheme in Italy – and through initiatives, which the ECB supports, to revive European ABS markets.
This matters for monetary policy for two reasons. First, it provides a spare tyre, i.e. an additional channel through which monetary policy can be transmitted if the bank lending channel becomes impaired again in the future. Second, it reduces banks’ market power, which may tend to make bank lending rates more responsive to monetary policy. Indeed, recent research has shown that where banks in Europe face limited competition, and firms depend on them, financing constraints for SMEs have been found to be higher. 
Such a rebalancing of the financing mix of the euro area economy probably requires further regulatory action to acknowledge the emergence of new, high-quality capital market instruments, such as simpler and more transparent ABSs.
Let me conclude.
Depending on how it materialises, bank deleveraging has a potential to impede the recovery and make it more difficult for central banks to engineer an appropriate degree of monetary accommodation or, to the contrary, to support the transmission of monetary policy. Therefore, policy-makers need to set proper incentives to promote “good” deleveraging and to steer clear of “bad” or “ugly” deleveraging.
Monetary policy has a vital role to play here, notably by providing the necessary liquidity support. But ultimately, it is up to policy-makers in other domains to ensure that the banking system is restored to health, and becomes fit for supporting the recovery and the reallocation of production factors in the euro area. Here, the steps being taken towards a banking union and the comprehensive assessment of banks’ balance sheets are crucial. They will establish the conditions necessary for a transparent, competitive and stable banking sector. And they will help monetary policy to regain traction across the euro area.
One final remark: for all this to work, the banking union needs to be complete. We need not only a single supervisory mechanism, but also a single resolution mechanism (SRM) and single resolution fund (SRF) and, at a later stage, a single deposit insurance scheme. If we fail to establish the SRM and SRF, bank resolution will remain a national task, resulting in a misalignment of responsibilities, and entrenching the link between banks and sovereigns. Not moving promptly towards the SRM and SRF would prolong uselessly financial fragmentation, leave the euro area financial system exposed to systemic fragility, and, ultimately, be harmful for growth and jobs. This would in particular be the case if the SRF is only slowly mutualised, and if it cannot resort to a common European backstop from the outset.
I thank you for your attention.
I would like to thank Federic Holm-Hadulla for his contributions. I remain solely responsible for the opinions contained herein.
See, for example, Albertazzi, U. and M. Bottero (2014), “The Procyclicality of Foreign Bank Lending: Evidence from the Global Financial Crisis.” Journal of International Economics, forthcoming.
Although the identification of “evergreening” practices is challenging, a large number of analyses have been conducted on this issue looking at Japan’s “lost decade”. See, among others, Caballero, R., T. Hoshi and A. K. Kashyap (2008), “Zombie Lending and Depressed Restructuring in Japan.” American Economic Review, Vol. 98, No. 5. One more recent study looks at the global financial crisis and identifies lending patterns consistent with evergreening practices, although limited to smaller banks; see Albertazzi, U. and D. Marchetti (2010) Credit supply, flight to quality and evergreening: an analysis of bank-firm relationships after Lehman, Bank of Italy Working Paper No. 756.
Euro area bank lending survey, ECB January, 2014.
Ryan, R., M. O’Toole and F. McCann (2014), “Does bank market power affect SME financing constraints?,” Journal of Banking & Finance.