DEPOSIT INSURANCE (DI) AS AN UNCOORDINATED INTERACTION Is harmonization sufficient? Theo Kiriazidis * Head of Research Department Hellenic Deposit and Investment Guarantee Fund (TEKE) * The usual disclaimer remains very necessary September 30 th, 2016
Scope The role of DI on cross border banking competition Despite the integration process, differences in certain DI conditions influence cross border deposit (capital) flows. Given certain DI conditions, States have an incentive to attract deposits. Certain DI conditions create antagonism with repercussions for financial stability and economic welfare 2
Empirical Evidence: Prior to the crisis In 1994 the 94/19/EC DGSD set a minimum cover limit of 20,000 with no upper limit. Competitive distortions emerged by coverage level differences Banks branches in host countries were covered by the home DGS. Depositors could promptly shift deposits amongst local branches and thus amongst DGSs. Nevertheless, most countries opted for the lowest coverage level Notable exceptions: France ( 70,000) and Italy ( 100,000). Prior to the crisis, the mass of deposits were essentially uninsured. 3
Empirical Evidence: Following the financial crisis (I) Substantial increase in DI levels 30 Sept. 2008 IRELAND 3 Oct. 2008 UK 5 Oct. 2008 GERMANY 5 Oct. 2008 AUSTRIA 6 Oct. 2008 DENMARK Government proclaimed full DI in biggest 6 domestic banks. Justification international financial crisis and uncertainty generated. Government announced rise in coverage level from 35,000 to 50,000. Prior level, 35,000, adopted on 1st Oct. following the Northern Rock run. Significant deposit outflow to Irish banks stressing UK banks in Northern Ireland. Government declared the full guarantee of all domestic banks deposits (political commitment). Germany s shift was induced by large electronic deposit flows during the weekend. Government announced the full guarantee of all domestic banks deposits. Justification to avert the flow of Austrian savings to Germany. Government announced the full guarantee of all deposits in Danish banks. Prior coverage level: 300,000 Kroner, roughly 40,000. 4
Empirical Evidence: Following the financial crisis (II) 7 Oct. 2008, EcoFin Council meeting in Luxembourg The EU finance ministers decided to: Increase the minimum guarantee level from 20,000 to 50,000 for one year, and Set the framework for 2009/14/EC DGSD adopting a harmonized level at 100,000 since Jan 2011 5
Questions Why low DI level is an initial policy equilibrium? Why a country would unilaterally increase its DI? Why other countries react? What is the outcome? 6
1. Depositors: Risk intolerant Actors (I) In absence of Deposit insurance: If their claims risk increases, withdraw some of their deposits, contributing to: banks de leverage and market discipline, or even a bank run, if the risk increase is sufficiently large and swift. In case of Deposit insurance: Depositors incentive shifts from the selection of banks to the selection of DI jurisdictions Market discipline is removed 7
2. Banks: Highly competitive Actors (II) The probability of banking failure inversely related and vastly sensitive to the level of domestic deposits. Attracting deposits from abroad enhances banking stability domestically Losing deposits abroad compromises banking stability domestically 3. States: Seek to protect their banking systems from destabilizing capital flights 8
Why low DI level is an initial policy equilibrium? DI reduces the severity of bank losses. Yet, it involves a funding cost (from private or public sources) and generates moral hazard Thus, in periods of financial stability it implies a net cost to economic welfare and states opt for unworthy deposit insurance This resembles the initial situation in the EU: DI levels sufficiently low ( 20,000) to be financed by the residual value of bank assets, without recourse to additional funds. The euro and the financial crisis changed the economic landscape. The euro has increased international bank competition The crisis made countries more sensitive to capital flows. 9
Why a country would unilaterally increase its DI? Country A increases its DI: to retain the level of domestic deposits following an economic shock (It is a means of fending off external shocks, a defensive response that unintentionally affects other States) 10
Why other countries react? Following COUNTRY A action, COUNTRY B experiences or foresees deposits (capital) outflows destabilizing domestic banking system. Foreign deposit insurance functions as a negative externality. COUNTRY B raises the level of DI to a point to stabilize its banking system i.e. to retain all domestic deposits. Country B s optimal reaction depends on certain factors: (1) the stress of the domestic banking (2) the proportion of willing to shift depositors (3) the transaction cost (4) the cost of providing DI (5) the uncertainty 11
The role of Uncertainty Uncertainty about the level of certain factors contributes to higher guarantee levels. Policy miscalculation of setting DI up to a sufficient level to avert destabilizing capital flows and avoid a crisis, is extremely costly, while the cost of additional DI is relatively little. 12
What is the outcome? The role of DI on cross border banking competition The DI level is a key policy variable affecting the location of deposits. External pressures determine significantly the provision of national DI. The looser the coordination, the higher the DI level Moral hazard Adverse selection Differences in coverage levels lead, not only to flight of deposits from low guarantee countries to higher guarantee countries, but also from lower guaranteed banks to higher guaranteed banks, which nevertheless operate within the same market. 13
TEMPORARY HIGH BALANCES (THB) COVERAGE: Another example of non harmonization The 2014/49/EU DGSD made significant progress towards harmonization. But provided that depositors Temporary High Balances (THB) emerging from certain transactions (such as real estate) should be protected by higher coverage levels which should be set taking into account the living conditions in the MSs. Member States should decide on a temporary maximum coverage level for such deposits and, when doing so, they should take into account the significance of the protection for depositors and the living conditions in the Member States. (Recital 26 of Directive 2014/49/EU) 14
Table 1 Temporary High Balances Coverage Level in selected Member States Countries THB Coverage Countries THB Coverage AT Austria 500,000 IE Ireland 1,000,000 BE Belgium 500,000 LV Latvia 200,000 CY Cyprus 50,000 LT Lithuania 300,000 FI Finland Full MT Malta 500,000 FR France 500,000 PT Portugal Full DE Germany 500,000 SI Slovenia Full EL Greece 300,000 ES Spain Full HU Hungary 50,000 UK United Kingdom 1.000.000 (GBP) Source: Council Ad Hoc Working Party on Strengthening of the Banking Union, May 2016. 15
Table 2 GDP per capita in Purchasing Power Standard (PPS) in Eurozone in 2014 Countries Euro Countries Euro Austria 35,500 Italy 26,400 Belgium 32,500 Latvia 17,500 Cyprus 22,400 Lithuania 20,600 Finland 30,300 Malta 23,600 France 29,300 Netherlands 35,900 Germany 34,500 Portugal 21,400 Greece 19,900 Slovenia 22,600 Ireland 36,800 Spain 25,000 Source: Eurostat Newsrelease, February 2016 This indicator can be used as a proxy to the living conditions. PPS takes into account differences in national price levels and allows meaningful volume comparisons of income ultimately available to private households in various countries. 16
Diagram 1: THB coverage and GDP per Capita for selected countries GDP per capita (euro) 40.000 THB Coverage 1.200.000 35.000 1.000.000 800.000 30.000 600.000 25.000 400.000 20.000 200.000 15.000 C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 C16 C17 0 GDP per capita for 2014 (PPP) THB Coverage Countries GDP per capita are set in ascending order, while their respective THB coverage levels are presented with red dots. For data analysis reasons full THB coverage is assigned the highest THB 17 coverage numerical level i.e 1,000,000.
THB coverage levels and GDP per capita follow completely different patterns. Some countries with lower GDP per capita have higher THB coverage levels. Running a simple statistical test: The correlation coefficient between THB coverage levels and GDP per Capita is 0.08 indicating essentially no correlation between the two indicators. 18
How did Member States decide the THB coverage levels? A cluster analysis: Selected countries are categorized in clusters according to THB coverage levels. Intra cluster competition is assessed. The degree of competition corresponds to cross border banks operation via branches 19
Table 3 THB Coverage Level and Cross Border Operation via Branches in Selected Member States Intra cluster competition Clusters Countries THB Coverage Level Number of branches from other countries within the cluster as a percentage of total Eurozone branches in Host country Ranking Cluster A Cluster B Cluster C C1 11.7% 3 Full coverage C2 27.3% 1 C3 36.4% 1 1,000,000 C4 7.2% 7 C5 56.6% 1 C6 51.4% 1 C7 500,000 38.7% 1 C8 61.5% 1 C9 71.4% 1 20
THB coverage determination THB coverage levels correspond to the structure of cross border branching. THB coverage levels are determined non cooperatively. Outcome towering DI levels 21
Some reflections Is the current situation in equilibrium (stable)? What happens if a bank from a high coverage country opens branches in a low coverage country? Will the latter not react by raising its coverage level? If yes, can it afford to? Should a bank s policy expansion determine the coverage level in a certain country? What about moral hazard and adverse selection? Can an efficient bank from a low coverage country expand via branches unhindered in a high coverage country? Is it a barrier to entry? Is coordinated equilibrium preferable to non coordinated equilibrium? 22
IS HARMONIZATION ON DI SUFFICIENT? In reality there are several aspects of DI (such as credibility, strength of backstops) that may affect deposits safety and thus deposit flows. DI harmonization cannot: avert antagonism eliminate perceptions of explicit full deposit insurance linked with home sovereign and thus reduce moral hazard prevent market fragmentation and competitive distortions. DI Mutualization? 23
The EDIS PROPOSAL According to the ECB, EDIS would enhance depositors confidence, and through risk diversification reduce the cost of DI. Thus it is beneficial from a pan EU point of view. MSs reactions to the proposal was mixed. Such divergence reflects differences in the respective payoffs EDIS may become part of an interaction within the EU. 24
Thank You!