Non-performing loans in the euro area – where do we stand?

Speech by Andrea Enria, Chair of the Supervisory Board of the ECB, at the Conference “EDIS, NPLs, Sovereign Debt and Safe Assets” organised by the Institute for Law and Finance, Frankfurt, 14 June 2019

By 2014, non-performing loans (NPLs) amounting to almost €1 trillion had piled up on the balance sheets of large banks in the euro area. That was a big problem. Since then, this amount has almost halved, and now stands at €580 billion. In the same period, the ratio of gross NPLs to total loans dropped from around 8% to 3.8%, falling below the 5% attention threshold defined by the EBA. Still, the ratio is above pre-crisis levels and significantly higher than in other major industrialised economies.

So, is the problem solving itself? No, it is not. Policy initiatives have played, and will continue to play, a key role in pushing banks to clean up their balance sheets.

The ECB has launched a few initiatives targeting NPLs, which have been fairly successful. We started with some general guidance to banks on how to deal with NPLs. Then we issued an addendum to this guidance which set out what we expect in terms of provisioning for new NPLs. And we later clarified what we expect in terms of provisioning for the stock of NPLs.

All these initiatives have taken a bank-by-bank approach, of course. Our expectations are bank-specific and strictly supervisory; they are not legally binding. They rather serve as a starting point for supervisory dialogue with each bank. Based on this dialogue, we might then adjust our expectations. All this happens within the second pillar of the Basel framework, our supervisory review and evaluation process (SREP).

In a sense, our initiatives have led the way. European legislation has recently been amended to include rules on minimum loss coverage for NPLs. These rules are legally binding and apply to banks across the board. So our initiatives are working well and have now been mirrored in minimum requirements embodied in actual law – Pillar 1. This makes the overall framework even more robust. We will now have to adjust our guidance to make sure that it is consistent and aligned with the new minimum requirements on NPLs, and that is what we are working on. But after these minor tweaks, I would say that the rules and supervisory policies to deal with NPLs are in place.

The problem of NPLs is not solving itself – and it has not yet been resolved. While it is true that the amount of NPLs has fallen significantly – by almost 50% since 2014 – the stock of NPLs is still very high. It is also very old. Many of the NPLs that we see on banks’ balance sheets have been there for years. For those banks with the highest levels of NPLs, more than half of their NPLs are older than two years and more than a quarter are older than five years.

At the same time, it seems that inflows of new NPLs are still on the high side – not least when you consider where we are in the business cycle. It also seems that some banks with high NPLs are still reporting increasing default rates. We find this somewhat worrying, and we urge banks to stem this inflow by rethinking their underwriting standards and engaging with distressed debtors.

So we have to continue our work in this area. When I say “we”, I am including the banks, of course. There are many ways to reduce NPLs. Banks can restructure viable borrowers, write off unrecoverable loans or sell them, for instance.

Originally, many authorities and banks were concerned that markets for selling NPLs might be shallow and illiquid, and dominated by just a few players. It seems, though, that supply has spurred demand and that the market for NPLs is now less of a buyer’s market than many had feared. NPL markets have become more active and have spread across the euro area. Between 2016 and 2018, we saw more and more transactions, sellers and buyers. In 2018 alone, banks from across the euro area sold or securitised around €150 billion of NPLs. Over the same period, they sold around €30 billion of foreclosed assets. For significant institutions with high levels of NPLs, sales and securitisations amounted to around one-third of NPL outflows in 2018.

So, more liquid and efficient markets do play a very important role in solving the problem of NPLs. But, in general, the very first thing banks should do is engage early with borrowers in trouble; they need to identify those borrowers who can be sustainably restructured. These are their customers and they are their responsibility. The earlier banks contact borrowers, understand their financial situation and deal with the issues head on, the better. The problem will not go away if banks turn a blind eye to it, do not invest adequate resources in resolving it or fail to develop the frameworks needed to manage customers.

We have to get a handle on this problem. We have to solve the issue of NPLs while the economy is still resilient. If banks have to sail into the next storm with too many NPLs on their balance sheets, they will be less able to weather it and come out safely on the other side.

Those banks with high levels of NPLs do understand this, and are acting accordingly. In 2018, many of them reduced NPLs by more than they had planned to. And looking at the most recent NPL strategies, their plans for the future are also quite ambitious. Most importantly, they are particularly aggressive in dealing with the older vintages of NPLs. These strategies, by the way, are a formal part of the SREP, and our supervisory teams closely monitor and challenge the progress banks make. In 2018, many banks overshot the targets by more than 25%. The problem of NPLs will not solve itself, but the banks can solve it – with our help.

Thank you for your attention.