Remarks given at IADI conference on "Designing an Optimal Deposit Insurance System"
Keynote address by Mr Stefan Ingves, Chairman of the Basel Committee and Governor of Sveriges Riksbank, at the IADI Conference on "Designing an Optimal Deposit Insurance System", Basel, 2 June 2017.
As prepared for delivery
Good morning, it is a great pleasure for me to participate in IADI's biennial research conference.
I have spent a large part of my professional career dealing with financial crises. While the characteristics of each of these crises differ, they all share a common theme. More specifically, the social costs stemming from them have been very large. These include direct costs, most notably fiscal costs from government intervention. But, more importantly, they also include broader costs to the real economy in the form of a contraction in lending, an increase in unemployment, and a reduction in economic activity and productivity.
Those costs point to the importance of building a strong safety net to safeguard financial stability and reduce the likelihood and magnitude of crises. One of the lessons I learned over the years is that the presence of a strong safety net is key to restoring confidence. In that regard, both the academic literature and practical experience from many countries suggest that deposit insurance systems play an essential role in protecting consumers and reducing the likelihood of bank runs.
While deposit insurance systems are a key ingredient to financial stability, their design differs vastly across jurisdictions. So the theme of this research conference - designing optimal deposit insurance systems - is of great interest and relevance from both an academic and public policy perspective. While an optimal framework may be difficult to achieve in reality, it is important to identify this endpoint and strive to get as close to it as possible.
In my remarks this morning, I will outline three challenges currently faced by deposit insurance systems. I will raise more questions than provide answers. But, before doing that, let me start by discussing some of the post-crisis improvements to global financial policy that are of particular relevance to this discussion.
Policy measures to strengthen the global financial system
The global financial crisis (GFC) highlighted a number of shortcomings in the design of financial policy. These fault lines included insufficient measures to curb excessive leverage, ensure robust levels of high-quality capital and liquidity, and mitigate the risks associated with too-big-to-fail financial institutions.
In the aftermath of the GFC, policymakers worked closely to address these shortcomings head on with a wide range of initiatives. For example, the Basel Committee on Banking Supervision revamped its regulatory framework to enhance the resilience of banks and to reduce the likelihood and magnitude of future banking crises.
Similarly, the Financial Stability Board developed a range of principles and initiatives related to the orderly resolution of global systemically important banks. In contrast to the public "bail-outs" of the past, the new resolution regime foresees a "bail-in" of large failing banks. An important objective of the new bail-in regime is that shareholders and creditors - rather than taxpayers - should bear the losses resulting from failure. Many countries are applying the same concept to a wider range of banks that are considered systemic for the national economy. This is changing the nature of the prudential safety net.
The interaction between the resolution framework and deposit insurance is a key challenge
With this background in mind, the first key challenge in developing optimal deposit insurance systems that I would like to raise today is:
What role should deposit insurance systems play in a world with bail-in tools?
In principle, the existence of bail-in tools means that there is an extra "cushion" of protection for insured depositors. Consequently, the likelihood of actual pay-outs from deposit insurance should fall. This raises (at least) three further questions in considering the design of optimal deposit insurance systems:
The first is whether the development of resolution regimes would have consequences for the fee structure of deposit insurance schemes. For example, for deposit insurance systems with risk-based levies, should a lower likelihood of pay-outs to depositors of systemically important banks result in lower fees? And should the fees paid by systemically important banks relative to other institutions change as a result?
A second question is the actual role of the deposit insurance system in the resolution process. Structurally, this differs currently across jurisdictions. For example, in the United States the FDIC has been responsible for both deposit insurance and resolution for a long time. In some countries, deposit insurers have been given stronger resolution mandates after the GFC. Other countries have set up, or are in the process of setting up, separate resolution funds in addition to the deposit insurance fund. So how should such funds best interact? When and how should each fund be used? Are there overlaps and how should the relevant authorities coordinate?
Furthermore, if a systemically important banks fails - and the losses are so large that the extra cushion is insufficient and the majority of a bank´s balance sheet becomes bailed-in - the deposit insurance scheme may have to replace the depositors and contribute to absorbing the losses. This could result in the deposit insurers´ claim being converted into equity and the deposit insurer becoming a new bank shareholder. This raises the issue of where the deposit insurance systems should be in the bail-in stacking order. Or whether legal provisions should exclude deposit insurers from a bail-in altogether. It also raises the issue of how deposit insurance systems should prepare themselves for potentially becoming bank owners.
A third question is how the introduction of resolution regimes affects market discipline and moral hazard. If one opts for relatively lower fees for systemically important banks, would that affect public confidence in the deposit insurance systems? What are the net effects of a stronger role for bail-in and a potentially reduced role for deposit insurance on market discipline, public confidence and moral hazard? Ultimately many of the answers to these questions depend on the effectiveness of the bail-in frameworks. However, the bail-in mechanism remains largely untested, and so is its interaction with deposit insurance. A number of questions still need to be answered.
Other challenges involve the relation between deposit insurance schemes and public guarantees
The interaction with the resolution framework is one. A second concerns the relationship between deposit insurance systems and public guarantees. To put it bluntly:
Are private deposit insurance systems as viable as public systems?
Deposit insurance systems have changed since the GFC. In 2009 the Core Principles for Effective Deposit Insurance Systems were developed jointly by IADI and the Basel Committee. One important principle aims at strengthening funding arrangements by stating that funding should be provided by member banks on an ex ante basis. Indeed more systems are now ex ante-funded than before the crisis.
Still, deposit insurance schemes are organised in very different ways around the world; some are private and others are public. They vary in the amounts covered and types of coverage. However, what they have in common is that they are essentially underfunded. That is, they are not intended to be able to fully cover the magnitude of a full systemic crisis. In fact, the target size of the funds generally only corresponds to a fraction of total insured deposits. If more money is needed in a country, the insurance scheme is ultimately backed by the state. These are important questions to consider. Can a private scheme be as credible as a public scheme? Can the private scheme provide public confidence in the same way? How do private deposit insurance schemes interact with public resolution schemes in times of stress?
Cross-border issues are another challenge
The third key challenge I would like to expand upon involves dealing with cross-border banks.
How should deposit insurance systems deal with cross-border banks?
Cross-border banking places considerable demands on cooperation between the authorities of different countries. However, even in normal times, it is not always easy to reach agreement. And it does not get any easier in a crisis.
Conflicts of national interest can be particularly destructive. There are situations where taxpayers in one country could risk ending up bailing out depositors of another country. This can lead to uncoordinated actions, possible value destruction and heated diplomatic disputes.
I suspect that many of you recall the experience of the Icelandic deposit insurance fund in 2008, which was unable to make good on its guarantees to over 300,000 primarily British and Dutch depositors who had placed their savings in IceSave, a branch of the Icelandic bank Landisbanki. In response, the British government attempted to freeze the bank´s assets in the UK using anti-terrorism legislation. Iceland, having too large a banking sector in relation to its economy to be able to back up the deposit insurance, ended up in an IMF programme, with far-reaching consequences for the Icelandic taxpayers.
Various arrangements exist when it comes to the obligations of home and host countries of a banking group with regard to both supervision and deposit guarantee. This is an area where the structure of the failing bank has important implications for the deposit guarantee scheme, and where the perspective of home and host countries can differ significantly when it comes to the treatment of "foreign" depositors.
Consider the case of a national deposit insurance scheme where a branch structure means that coverage extends to depositors in the host country. The question of whether a bank has a subsidiary or a branch structure is then of obvious importance to the home country. From a host country perspective, the question is whether the home country government will back the deposit insurance guarantee. It is entirely reasonable to expect that the capacity of some governments to provide for the guarantee may be questioned, rightly or not.
Some of the challenges I have discussed concerning the financial safety net for cross-border banks can be greatly affected by changes in the way banking groups are organised. How will the safety net be affected and what consequences does that have for the relevant authorities, when, for example, a decision is made to move headquarters to another country, or when there is a merger between banks in different jurisdictions?
Let me give you an example. The Swedish banking group Nordea, one of the 30 Global Systemically Important Banks, recently switched from a subsidiary to a branch model.
This means that large parts of Nordea´s foreign banking operations, which were previously conducted via its foreign subsidiaries, are now conducted through branches. According to regulations pertaining to our region, such changes not only entrust the Swedish authorities with greater responsibility for both supervision and crisis management of the whole company, but it will also considerably increase the responsibilities of Sweden as home country.
Such responsibilities include possible liquidity support from the Riksbank and potential public crisis management measures, if deemed necessary. The Nordea restructuring also means that deposits in the group, which are covered by the Swedish deposit insurance scheme, have increased significantly. If the deposit insurance scheme is ultimately backed by the state, then the Swedish state will back, de facto, Finnish or Latvian depositors, for example. What will be the reactions of Swedish taxpayers in a crisis situation? And will foreign depositors trust the Swedish deposit guarantee scheme and, ultimately, the Swedish state to treat them fairly? Will there be sufficient economic resources to do so?
When it comes to possible liquidity support and the deposit guarantee scheme, currency mismatches may also be an issue. Both Finland and Latvia use the euro, while the Swedish deposit guarantee scheme is funded in Swedish krona. Such a mismatch could be challenging in some crisis scenarios (in particular, if swap markets were to be dysfunctional at the time).
I believe that cross-border integration is likely to favour overall economic growth and welfare. But when considering changes to banks' organisational structures, it is also important for the authorities to discuss the implications that this will have for the financial safety net and to make the right preparations. In my opinion, changes in the operations of cross-border banking groups, such as "branching", should therefore be met with stricter requirements on the bank to maintain liquidity buffers in all relevant currencies. It should also trigger a discussion about the relevant size and composition of central bank foreign exchange reserves and about the deposit guarantee fund, as well as a discussion about possible swap agreements among central banks. In this regard, cooperation among deposit insurers is also critical.
To sum up
As I have just mentioned, there are several challenges related to developing optimal deposit insurance schemes.
- First, the interaction between the deposit insurance scheme and the resolution framework.
- Secondly, the relationship between deposit insurance schemes and public guarantees.
- And, third, issues relating to cross-border banking.
However, even if designed optimally, I believe that an effective deposit insurance scheme should only be drawn upon rarely. This highlights the importance of enhancing the resilience of financial institutions in an upstream fashion, most notably for banks, to reduce the likelihood of financial crises.
To that end, the post-crisis reforms developed by the Basel Committee - ranging from higher levels of capitals, international liquidity standards, macroprudential measures to more robust risk capture - play an important role when considering the optimal design and functioning of deposit insurance systems. Indeed, on balance, we should place more weight on ensuring that the ex ante prudential regulatory framework is sufficiently robust in its design and calibration, so as to limit the likelihood of having to rely on deposit insurance schemes.
My final point is to remember that financial stability is a joint responsibility. The financial system safety net as a whole - not just its individual components - must be sound and resilient.