Two reports published in early 2020 shed new light on this challenge. The European Central Bank’s risk report on less-significant institutions is the first of what is intended to be an annual series. The impact assessment study on the most important differences between accounting standards used by banks in the banking union was prepared by legal consultants for the European Commission’s Directorate-General for Financial Stability, Financial Services and Capital Markets Union (DG FISMA). The reports provide comparative quantitative information that was not previously available in the public domain.
The new data highlights differences in rules and oversight in different countries that matter in the context of efforts to achieve an EU single market and banking union. From a single-market perspective, the fact that many German banks are subject to a different supervisory, state-aid and accounting framework raises the possibility of competitive distortions, even though a subset of these banks only have local activity and don’t compete outside of Germany. German stakeholders might misjudge and underestimate the extent of EU-imposed discipline in the banking sectors of other member states, by wrongly assuming that national discretions and exceptions in those countries are similar to those in Germany. Conversely, in euro-area countries other than Germany, a politically corrosive perception of unfairness can arise from differences in regulatory and/or supervisory treatment. From a banking-union perspective, the exceptions could, at least in certain scenarios, contribute to fragmentation of the euro-area financial space and thus to the risk that bank-sovereign vicious circles become worse, and potentially unmanageable.
The history that led to this situation is too long to summarize here. In a nutshell, the banking systems of a number of euro-area countries previously displayed significant idiosyncrasies, memorably expressed by former prime minister Giuliano Amato’s description of Italy’s banking system in 1988 as a “petrified forest.” But in Italy as in other euro-area countries, such idiosyncrasies have been eroded by successive waves of reform. Germany has reformed less than most others, largely because it has not been compelled to by circumstances. It also has unique leverage over EU legislation, which could have played a role in the creation or persistence of legislative loopholes. Furthermore, Germany’s public bank community is uniquely intertwined with its political community.
The ECB risk report gives, for the first time, a national breakdown of the assets of significant versus less-significant institutions in the euro area. This distinction matters for the assigning of supervisory authority within the banking union. Significant institutions (SIs) are all euro-area banks with more than €30 billion in total assets, plus a few more on the basis of criteria that include cross-border activity and prominence within a single country. Less-significant institutions (LSIs) are all the other banks. Only a small number of banks are deemed SIs by the ECB for reasons other than total assets above €30 billion (31 in the ECB’s latest listing, compared to more than 2,600 LSIs), so that the boundary between SIs and LSIs is primarily, though not exactly, determined by the asset-size criterion. The ECB directly supervises SIs, while most supervisory tasks related to LSIs are carried out by national supervisory authorities (though the ECB has authority over banking licences and other infrequent procedures). The ECB also exercises oversight over the national supervision of LSIs.
A fact that jumps out from the report is that LSIs are found disproportionately in Germany (Figure 1). At end-2018, German LSIs represented 55% of total LSI assets in the euro area, whereas Germany accounted for 25% of total banking assets (LSIs and Sis).
Figure 1: Aggregate assets of SIs (ECB supervision) and LSIs (national supervision), end-2018 (€ bns)
Source: ECB risk report on LSIs, Table 1; IPS (institutional protection schemes, see below) assets based on The Banker database, Deutsche Bundesbank (page 110) for Germany, and Grunewald 2017 (Figure 1, page 8) for Austria and Spain.
The vast majority of German LSIs, however, are not fully on their own, even though they are managed on a decentralised basis. They benefit from mutual support arrangements known in EU prudential law as institutional protection schemes (IPSs), labelled “virtual groups” by the ECB’s then second-most-senior supervisor at the start of her tenure in 2014. LSIs within an IPS support each other: if and when one becomes unviable, it is generally rescued by its peers. As a consequence, the group as a whole, rather than its individual member banks, is the relevant level of observation for financial-stability purposes. As Figure 1 shows, not all LSIs belong to an IPS. Conversely, all existing IPSs include at least one SI.
There are two IPSs in Germany: the Savings Banks Financial Group (Sparkassen-Finanzgruppe) and the Volksbanken Raiffeisenbanken Cooperative Financial Network (Genossenschaftliche Finanzgruppe Volksbanken Raiffeisenbanken, hereafter ‘Cooperative Group’). Entities in the Sparkassen-Finanzgruppe belong to the public sector and are controlled by different sub-federal levels of government under various legal forms and ownership patterns. Such entities include local savings banks (Sparkassen) and regional wholesale banks (Landesbanken). Entities in the Cooperative Group are ultimately owned by the individual cooperative members.
As of end-2018, based on data from the German Central Bank, LSIs in the Sparkassen-Finanzgruppe represented 45% of total German LSI assets. Those in the Cooperative Group, including buildings and loan associations, represented another 39%. In recent years, both groups have published group-level financial statements (using the accounting methods, respectively, of aggregation and consolidation), which indicate that, taken together including their SI members, they are among the largest banking players in the euro area. Indeed, in 2018 the Sparkassen-Finanzgruppe in aggregate had more assets than any euro-area bank (Figure 2).
Figure 2: Total end-2018 assets of the largest euro area banking groups (€ bns)
In black: Group-level assets of German institutional protection schemes; in red: other German banks; in yellow: other banks with total assets above €1 trillion. Source: The Banker database, Savings Banks Group website, and Cooperative Group website.
On the basis of publicly available information, it is difficult to assess the specific differences, if any, between the supervisory regimes of the ECB (applicable to SIs, including those in IPSs) and of national authorities (applicable to LSIs, including those in IPSs). The applicable prudential rulebook is substantially harmonised by the EU capital requirements directives and regulations. Thus, differences which may remain despite the above-mentioned supervisory oversight by the ECB, are mostly in supervisory enforcement and discretion. There are indications that at least some banks have a preference for being labelled LSIs. For example, L-Bank (full name Landeskreditbank Baden-Württenberg Förderbank), a public bank in southern Germany, unsuccessfully sued the ECB over its SI determination. It was later removed by new legislation from the scope of application of EU banking law altogether (together with two other former German public SIs, NRW.Bank and Landwirtschaftliche Rentenbank), and is thus no longer under direct ECB supervision.
European banks are subject to state-aid control conditions, enforced by the European Commission. The application of these disciplines, however, takes a distinctive form for unlisted public banks, for which the boundary between an arm’s-length recapitalisation (with the government providing funds as a shareholder) and state aid (the government transferring funds in a non-market transaction) is less well-defined than in cases when there are shareholders other than public entities. The leeway that results from full government ownership was put under the spotlight in the recent case of Norddeutsche Landesbank (or NordLB), a public bank in Northern Germany, which was recapitalised by its public shareholders in a transaction that the European Commission deemed “market conform” in December 2019 but was widely viewed as distortionary by external observers. The NordLB decision itself had precedents, particularly in the cases of Portugal’s Caixa Geral de Depositos in 2017, and Romania’s CEC Bank in October 2019. The upshot is that the state-aid regime appears to be different for unlisted public banks than for other banks, in practice if not in theory.
Possibly for the first time, the ECB’s report on LSIs, combined with data on assets of individual SIs and a few no-nonsense assumptions, permits a tentative mapping of such banks in the euro area. Figure 3 shows the results of these calculations. They suggest that, as with LSIs, unlisted public banks in the euro area are predominantly located in Germany. Namely, the Sparkassen-Finanzgruppe includes all of Germany’s unlisted public SIs, and most if not all of its unlisted public LSIs. (Note: the differences compared to Figure 1 and Figure 4 in total amounts by country result from differences in accounting methodologies and in the scope of observation, eg SIs that are part of non-euro-area banking groups are not included in Figure 3, and some cross-border operations are assigned to the home country in Figure 3 vs host country in Figures 1 and 4).
Figure 3: Unlisted public banks in the euro area, end-2018 assets (€ bns)
Source: ECB risk report on LSIs; Deutsche Bundesbank; The Banker database; corporate websites; author’s assumptions and calculations.
In landmark legislation adopted in 2002, the EU mandated the use of International Financial Reporting Standards (IFRS) for all its listed companies, starting in 2005. For unlisted companies, however, including unlisted banks, the choice of accounting standards was left to the discretion of individual member-state authorities. This stands in contrast to most of the world’s jurisdictions outside of the European Union, where all (listed and unlisted) banks generally have to comply with IFRS or, in the case of the United States, Generally Accepted Accounting Principles (US GAAP).
The DG FISMA report on banks’ accounting practices includes an overview of which banks (aggregated by country and by assets) use which set of accounting standards, – IFRS or national standards. Germany stands out with 52.1% of banking assets reported under national standards. The next-highest ratio is much lower, in Austria (22.7%), followed by the Netherlands (5%), and under 2.5% in all other member states. Figure 4 illustrates these findings.
Figure 4: Use of accounting standards by banks in the euro area, end-2018 assets (€ bns)
Source: European Commission impact assessment study, Table 1 (page iv) for percentages; ECB risk report on LSIs for total assets per country.
Figure 4 implies that there remain only two systems of accounting standards in wide use in the euro-area banking system: IFRS and German national standards. Whether one is more demanding than the other depends on the issue; the DG FISMA report gives a wealth of detail on the differentiated outcomes. For example, for credit loss accounting, IFRS require expected loss provisioning as mandated by the Group of Twenty (G20), while German standards preserve more flexibility to maintain the prior practice of incurred loss provisioning.
It is too early to assess the extent to which the COVID-19 pandemic may influence future debates and decisions on the still-unfinished banking union. By providing additional data on structural quirks of the euro-area banking system, the two reports analysed in this post contribute to a trend of greater supervisory transparency. Further efforts in that direction will hopefully contribute to better-informed policymaking when dealing with the difficult challenges ahead.