Journal Issue

Euro-Area Banking At the Crossroads

Agnes Belaisch, Joaquim Levy, Laura Kodres, and Angel Ubide
Published Date:
March 2001
  • ShareShare
Show Summary Details

I. Introduction

1. Euro area banking systems are in flux. Faced with the combined pressures of globalization, disintermediation, new technologies, and increased competition from nonbank financial intermediaries, banks are devising strategies to thrive in this new environment. These trends, while global, have recently affected the euro area with particular intensity. The structural change brought about by the adoption of a common currency and a common monetary policy is exerting a profound impact on the area’s financial sector. Money markets have been integrated; trading, quoting and settlement operations are already denominated in euros; and competition for banking services—at the wholesale level through global providers and at the retail level through internet banking—is intensifying. Further, multiple European currency trading desks are extinct and are being replaced by bond and equity trading and issuance. For these reasons, this paper focuses on the euro area to examine the challenges facing banks, their response, and that of their supervisors in this new environment. Both banks and those individuals that influence public policy will need to continue to be proactive in developing appropriate strategies to facilitate adaptation to the new euro area environment.

2. This paper is divided into two parts. Section II gives an overview of the euro-area2 banking system, identifying common trends in bank performance and balance sheet structure, and outlines some recent capital market developments. This provides the background for the discussion, in Section III, of the issues arising from these trends and developments. In particular, the following points are examined in detail.

  • Financial disintermediation has increased in the wake of Stage III of European Economic and Monetary Union (EMU). Is such a trend a threat to the profitability of euro-area universal banks? Or, being universal banks, do they have the flexibility to adjust to the changing environment?
  • The ongoing process of consolidation has been rationalized on the basis of the benefits of the economies of scale and scope it can generate, which can vary across lines of business. Are the conditions for realizing these benefits—free access to new markets and flexibility in the use of inputs, including labor—satisfied in the euro area?
  • Consolidation, especially at the retail level, has predominantly prevailed within a national context rather than across borders. Could this hurt competition? To what extent could it hamper the benefits of technological advances such as the Internet?
  • An outstanding feature of banking systems in the euro area is the large share of financial institutions whose ownership structures mean that they are “not for sale.” How could this feature affect the reshaping of the financial system in the region? Should these institutions be incorporated?
  • Considering that the “optimal” size of banks may have grown, do existing impediments to cross-border consolidation promote the creation of “too big to fail” institutions from a national perspective?
  • What are the implications of market integration, consolidation across markets and products, and creation of very large firms, for financial stability and the soundness of the banking system? Has the risk of cross-border contagion increased? The advent of Stage III of EMU implies that the domains of monetary policy and banking supervision no longer coincide. How does this impact the arrangements for the prevention and management of liquidity and solvency crises?
  • Is the current supervisory structure appropriate for this quickly changing environment, in particular as regards the development of very large, complex institutions and the growing regional integration of financial markets, foreshadowed by the rapid expansion of interbank credit following monetary union? Would a centralized supervisory architecture be more appropriate?

The answers to these questions are not straightforward, and in many cases depend crucially on whether a national or euro-area view is taken. The presence of cross-border interrelationships and linkages in an integrated economic area suggests that a euro-area view may be preferable. But even from a euro-area perspective, different views regarding the speed and desired amount of integration of political institutions will yield different answers. Overall, the paper takes the view that, to preserve the stability of financial markets and increase their efficiency, the changing euro-area banking environment requires policy makers to examine financial sector policies increasingly from a euro-area perspective, complementing the domestic one. On perhaps the most difficult question facing policy makers—the appropriate venue for financial oversight—the paper discusses the costs and benefits of various arrangements, noting that such discussions are already taking place in other fora (as evidenced by the Brouwer Report). More specifically, the paper concludes that a centralized euro-area supervisory organization, although perhaps desirable in principle, would be premature at this stage in practice. However, the increased integration of financial markets calls for more centralized market surveillance that builds on domestic supervisory information. This, together with enhanced cooperation of national supervisors, is key to the efficiency and stability of euro-area financial markets.

II. Structure and Performance of the Euro Area Banking Sector

A. Market Structure and Institutions

3. Although it is difficult to characterize the euro-area banking sector using just one or two phrases, it is clear that the euro area’s financial system continues to be bank dominated.3 The proportion of financial assets controlled by the banking systems of the euro area countries remains high (Table 1). Bank loans to euro-area residents amount to about 100 percent of the area’s GDP, twice the ratio in the United States, whereas equity and bond market capitalization as a percent of GDP are substantially smaller than in the United States. Compared to Japan, euro-area bank loans as a percent of GDP are similar, though Japanese banks rely more heavily on deposits. Like the United States, debt and equity markets play a larger role in the financial structure of Japan than in the euro area. While euro-area banks continue to play a dominant role in intermediating savings through the traditional means of collecting deposits and extending loans, the use of investment funds as well as pension and insurance products as savings vehicles is growing. For instance, assets in investment funds have increased at double-digit rates in the recent years in almost all countries. Loan growth to euro-area residents has risen at lower, albeit still brisk rates, on average about 7 percent in 1998-99. Contrary to the United States, where commercial banks have only recently been able to issue mutual funds and sell insurance products, the growth in new savings vehicles is replacing traditional revenue of euro-area universal banks: revenues are not necessarily lost to other financial intermediaries.4 This section identifies common features and trends, while pointing out some distinct differences, among EU-11 countries’ banking sectors. Subsection C will discuss recent performance and balance sheet trends in the sectors.

Table 1.Financial Structure in the Euro Area, the United States and Japan
Reporting PeriodUnitEuro AreaUnited

Bank deposits 1/June 1999EUR billion4,752.24,742.84,467.5
June 1999Percent of GDP77.855.2111.7
Bank loans 2/June 1999EUR billion6,136.14,154.84,280.8
June 1999Percent of GDP100.448.4107.0
Outstanding domestic debt securitiesJune 1999EUR billion5,422.714,140.85,061.1
June 1999Percent of GDP88.8164.6126.5
Issued by corporatesJune 1999EUR billion202.32,493.8583.4
Issued by financial institutionsJune 1999EUR billion1,891.53,900.1753.7
Issued by the public sectorJune 1999EUR billion3,329.07,746.83,723.9
Stock market capitalization 3/October 1999EUR billion4,346.013,861.16,275.8
October 1999Percent of GDP71.1163.3137.7
Sources: (From ECB Monthly Bulletin, January 2000) BIS (domestic debt securities); ECB (bank deposits and MFIs loans for the euro area); IMF (bank deposits and GDP forecasts in Japan and the United States); Eurostat (GDP forecast for the euro area, stock market capitalization); Federal Reserve (bank loans for the United States); and Bank of Japan (bank loans for Japan).

For the euro area, data cover demand deposits, deposits with an agreed maturity and deposits redeemable at notice of MFIs other than NCBs and the ECB. For the United States, data include demand, time and savings deposits from all banking institutions, while for Japan, data include demand and time deposits from deposit money banks. For Japan, other banking institutions which offer close substitutes to deposits are not included.

For the euro area, data cover loans to euro area residents from the consolidated balance sheet of euro area MFIs. For the United States, data cover total loans at commercial banks, saving institutions and credit unions (from the Flow of Funds accounts of the United States from the Federal Reserve), while for Japan data cover total lending at deposit money banks from the monetary survey from the Bank of Japan.

Owing to the use of different reporting rules and calculation methods, data are not entirely comparable.

Sources: (From ECB Monthly Bulletin, January 2000) BIS (domestic debt securities); ECB (bank deposits and MFIs loans for the euro area); IMF (bank deposits and GDP forecasts in Japan and the United States); Eurostat (GDP forecast for the euro area, stock market capitalization); Federal Reserve (bank loans for the United States); and Bank of Japan (bank loans for Japan).

For the euro area, data cover demand deposits, deposits with an agreed maturity and deposits redeemable at notice of MFIs other than NCBs and the ECB. For the United States, data include demand, time and savings deposits from all banking institutions, while for Japan, data include demand and time deposits from deposit money banks. For Japan, other banking institutions which offer close substitutes to deposits are not included.

For the euro area, data cover loans to euro area residents from the consolidated balance sheet of euro area MFIs. For the United States, data cover total loans at commercial banks, saving institutions and credit unions (from the Flow of Funds accounts of the United States from the Federal Reserve), while for Japan data cover total lending at deposit money banks from the monetary survey from the Bank of Japan.

Owing to the use of different reporting rules and calculation methods, data are not entirely comparable.


4. Unlike other developed countries, virtually all euro-area countries continue to maintain savings banks, and mutual and/or cooperative banks that wield considerable weight in the local market, particularly at the retail level (Table 2). Although in most EU-11 countries the largest institutions in the banking sector are private commercial banks, other types of banks with different ownership structures continue to play a substantial role in the banking sector. The definitions differ slightly from country to country, but these institutions can generally be characterized along the following lines: commercial (private-stock companies), savings banks, cooperative or mutually owned banks, public banks, and a mixture of other types of banks, usually with special purposes (Box 1). In several cases, the savings banks are publicly owned, most often by local or municipal authorities, and there are still cases of state and central government ownership of big banking institutions.

Table 2.Market Shares per Type of Institution in Selected Euro-Area Countries(End-1998, in percent)
In Percent of

Total Assets
In Percent of

Total Deposits
In Percent of

Total Loans
Commercial banks54.138.443.5
Savings and cooperative banks28.460.136.5
Commercial banks47.943.947.3
Savings and cooperative banks 1/27.850.233.4
Commercial banks55.748.652.7
Savings and cooperative banks38.748.140.2
Commercial banks81.183.285
Savings and cooperative banks 2/13.315.713.6
Source: Central Bank bulletins.
Source: Central Bank bulletins.

5. Germany has the largest banking system in terms of number of credit institutions and Ireland the smallest. Germany has by far the biggest number of credit institutions—over 3,000—due to its large population and the diverse nature of its banking system (Annex I, Table 2). France has fewer than half that many, followed by Austria and Italy (with each containing slightly fewer than 1,000). Spain ranks sixth after the Netherlands. Each of the remaining countries contains fewer institutions by far.

Box 1.Types of Euro Area Banking Institutions and their Roles

Private commercial banks: Commercial or private banks are owned by their shareholders. Such private-stock companies usually offer equity to the public, but may be owned by private equity holders. They can distribute profits to their shareholders, typically in the form of dividends. These owners generally have limited liability and exercise control through various mechanisms, often through boards of directors or supervisory bodies. Voting rights, though, may be separable from share ownership.

Savings banks: Savings banks often supply credit to local or regional areas. In many cases, their original purpose was to provide credit to farmers, artisans, or other underprivileged groups who were unable to obtain credit elsewhere. Even when not required to do so, savings banks often focus on individuals and small- and medium-sized businesses. And even when shareholder-owned, there may be restrictions on the source of deposits, choice of assets, or distribution of profits. When partly or entirely owned by state or local governments or municipalities, these institutions are usually required to allocate part of their operating surplus to a “social fund” for use in the local community and the remaining profit can be either retained or distributed to the government owner. Further, in many cases, government officials acting on behalf of the bank are involved in lending policies and day-to-day operations to assure that the bank is fulfilling its stated public purpose. The institution may also receive either subsidized capital, loan guarantees, or the ability to provide preferential interest rates to depositors (higher than market rates) or to borrowers (lower than market rates). Governance structures vary considerably, particularly when the savings institutions is partly or fully owned by the public sector.

Cooperative/mutual banks: These banks are typically owned by their depositors or creditors and the services of these banks may be restricted to those who own them, although recent liberalization has permitted many of these institutions to offer their services to others. Ownership shares can be restricted to ensure broad ownership. In some countries, profits are distributed as dividends to the mutual owners, sometimes in the form of higher interest rates on deposits. In other countries, profits are retained, adding to reserves and the equity base. Governance is often implemented through boards of directors selected from among the members of the cooperative or mutual institution.

Public banks: Public financial institutions are now less prevalent in Europe and are typically outside the banking system. However, the most common type of public banks remaining in Europe are savings banks, owned or controlled in part by local or municipal authorities. Germany and Austria, with 35 percent and 14 percent of assets in banks either owned or governed by the public sector, respectively, constitute the largest public banking sectors of this type.

Other types of banks: Often countries have some specialized lending institutions. For instance, many EU countries contain mortgage banks, Germany has the largest such sector, whose assets are predominately mortgages and their liabilities come from either household deposits or the issuance of mortgage-backed securities. In some countries, there are agricultural lending banks, postal savings banks, and other special banks servicing specific sectors of the economy.

Distribution channels and employees

6. Although the number of banks can be a useful starting point, perhaps a better measure of the availability of banking services to customers is the number of branches per 1,000 inhabitants (Table 3). By this measure, Spain has the highest branch “coverage” with nearly one branch for every 1,000 inhabitants compared with an average of 0.57 branches per 1,000 across all euro-area countries. The smaller populations of Luxembourg and Belgium also have a considerable number of branches per 1,000 inhabitants—0.92 and 0.70 branches respectively. Germany, despite its overwhelming number of banks, has 0.55 branches for every 1,000 inhabitants, ranking fifth of all euro-area countries’ banking sectors.

Table 3.Number of Bank Branches per 1,000 Capita
Spain 1/0.900.920.940.960.98
Luxembourg 2/0.780.850.920.950.92
United Kingdom0.380.300.280.280.27
United States0.
Sources: National central banks; national bankers’ associations; European Central Bank: “Payment Systems in the EU,” 2000; FDIC; and IFS.Notes: Includes foreign bank branches; excludes branch bureaus in other banks. Excludes post office branches, central bank branches, Treasury branches. Ireland and UK figures include branches of building societies.

Data in 1990 is for 1991.

Figures in 1998 exclude branches of foreign banks.

Sources: National central banks; national bankers’ associations; European Central Bank: “Payment Systems in the EU,” 2000; FDIC; and IFS.Notes: Includes foreign bank branches; excludes branch bureaus in other banks. Excludes post office branches, central bank branches, Treasury branches. Ireland and UK figures include branches of building societies.

Data in 1990 is for 1991.

Figures in 1998 exclude branches of foreign banks.

Table 4.Employees per Branch
Germany 1/14.214.614.815.5
Sources: National central banks; National Bankers’ Association; BIS; and ECB “Payment Systems in the EU,” 2000.

Post offices are not included.

Sources: National central banks; National Bankers’ Association; BIS; and ECB “Payment Systems in the EU,” 2000.

Post offices are not included.

7. While in most countries the number of credit institutions has been shrinking over the last 15 years (the exceptions are Ireland and Luxembourg, where the increased number is attributable to foreign banks engaged in cross-border activities), the number of branches per capita started its decline somewhat later. In fact, in 4 of the 11 euro-area countries, this measure of banking density continues to increase (Table 3). Both Spain and Italy are among the countries for which branches per capita have continued to expand—often through the split of large branches—even while the number of banks has continued to decline.5 Finland stands out as a euro country in which the branch network per capita has shrunk dramatically since 1985, though this is likely the result of the Nordic banking crisis in the early 1990s, which encouraged a dramatic restructuring in the Finnish banking system.

8. Taking the place of traditional brick and mortar branches are new distribution channels. The number of ATMs per 1,000 inhabitants has grown substantially in all EU countries, with Spain leading the pack (Table 5). The smaller countries that have experienced branch closings have seen some of the greatest usage of ATMs. More recently, telephone and Internet banking options have been introduced. For instance, up to 25 percent of Finland’s Merita bank customers already use Internet banking.6

Table 5.Number of ATMs per 1,000 Capita
United Kingdom0.250.360.380.400.42
United States0.470.520.620.69
Sources: National central banks; National Bankers’ Association; ECB, “Payment Systems in the EU,” 2000; and IFS.
Sources: National central banks; National Bankers’ Association; ECB, “Payment Systems in the EU,” 2000; and IFS.

9. Given the trends noted above, a declining number of bank employees as a proportion of the population would be expected, but only recently has such a trend been evident. The last several years have seen only marginal changes in the number of bank employees in most countries (Table 6). Moreover, even in cases where the number of employees has declined, labor costs have remained stubbornly high.7 This is probably due to the type of individuals now being employed (higher skilled) and high nonwage costs (pension and social insurance contributions, separation costs, etc.).

Table 6.Number of Employees
Germany 1/695,900757,800751,100752,550
Sources: National central banks; National Bankers’ Associations. Figures differ according to sources: data from central bank publications are reported when available.

Post office employees are not included.

Sources: National central banks; National Bankers’ Associations. Figures differ according to sources: data from central bank publications are reported when available.

Post office employees are not included.


10. To date, consolidation in the euro area has taken mostly two forms, (1) mergers among relatively large private, commercial banks and among bank and nonbank financial institutions; and (2) mergers within the savings and cooperative banks, respectively. Consolidation has been essentially limited, sometimes with implicit government guidance, to within national borders and within their own types (see subsection B below). Some commentators have interpreted the governments’ guidance as an apparent desire to limit foreign ownership of some influential institutions and to create a few “national champions” in each country to compete in the European or global marketplace (Table 7).

Table 7.Recent Mergers and Acquisitions Among Large Banking Groups
CountryBanking Group
SpainBSCH (Banco Santander + Banco Central Hispano+Banesto)

BBVA (Banco Bilbao Vizcaya + Argentaria)
AustriaBank Austria (Bank Austria + Creditanstalt)

Erste Bank (Giro Credit + Erste SparCasse)
ItalySanPaolo IMI (Instituto Bancario SanPaolo di Torino + IMI)

Banca Intesa (Banco Ambrosiano Veneto + Cariplo + CPP) + BCI

Unicredito Italiano (Credito Italiano + Unicredito)
GermanyHypo Vereinsbank (Bayerische Vereinsbank + HypoBank)

Deutsche Bank + Bankers Trust
FranceBNP-Paribas (BNP+ Paribas)

Banques Populaires + Natexis

Crédit Mutuel + CIC

Caisse d’Epargne + Crédit Foncier

Société Générale + Crédit du Nord

Crédit Agricole + Banque Sofinco + Banque Indosuez
PortugalBanco Comercial Portugues + Banco Portugues do Atlantico

Caixa Geral de Depositos + Banco Pinto & Sotto Mayor
BelgiumKBC (Kredietbank + Cera)

Bacob + Artesia Bank
NetherlandsABN-Amro (ABN + Amro)
DenmarkUnibank (Unibank + Tryg-Baltica)
Cross-borderDexia (Crédit Local de France and Crédit Communal de Belgique)

Fortis (Générale de Bank and ASLK-CGER Bank)


ING + Banque Brussels Lambert


BSCH + Totta & Acores

Bank Austria + Hypovereinsbank

BSCH-Royal Bank of Scotland-SanPaoloIMI-SG-Commerzbank-


Crédit Agricole-Crédit Lyonnais-Banca Intesa

BBVA-Banco di Napoli-BNL-Credit Lyonnais

ABN-Amro - Banca di Roma
Sources: Various company reports and financial press releases.
Sources: Various company reports and financial press releases.

11. Consolidation has accelerated recently at the top: more than half of the 30 biggest euro-area banks are the result of recent mergers and the average size of the top five has doubled since 1995 (Table 8). Bank of International Settlements (BIS) data show that some 500 mergers and acquisitions (M&As) took place in 1991-92, valued at $17.5 billion, whereas in 1997-99 only 200 M&As took place, at a value of about $100 billion—fewer M&As but of a much larger scale.

Table 8.The Top 30 Euro Area Banking Groups(In terms of total assets, in million of U.S. dollars)
Crédit Agricole241992Deutsche Bank368261Deutsche Bank + Bankers Trust*732534
BNP231463Crédit Agricole328152BNP Paribas*688361
Crédit Lyonnais210727Crédit Lyonnais327903ABN-Amro*504122
Deutsche Bank202263ABN-Amro290835Hypovereinsbank*504122
Société Générale164741Société Générale278006Crédit Agricole +Indosuez*455792
Caisses d’Epargne152722BNP271635Société Générale447545
Dresdner Bank147001Dresdner Bank253818Dresdner Bank427261
Paribas138668Paribas242219Westdeutsche Land Giro408372
Commerzbank112825Westdeutsche Land Giro237535Commerzbank381359
DG Bank109168Commerzbank220704ING Bank Group*326813
IBSan Paolo107403Bayerische Vereinsbank204423Fortis*323567
Westdeutsche Landesbank Giro104508Caisses d’Epargne187411Rabobank Netherlands291353
Bayerische Vereinsbank102191Bayerische Hypotheken & Wechsel Bank177540Crédit Mutuel + CIC*284461
BNL100967Bayerische Landesbank171816Bayerische Landesbank284064
Amro Bank93824Krediteanstalt fur Wiederaufbau158736BSCH*258872
Bayerische Landesbank90855DG Bank158227DG Bank248297
ABN90411Bankgesellschaft Berlin157197Crédit Lyonnais243708
Bayerische Hypotheken & Wechsel Bank90129Rabobank Netherlands155082Banques populaires + Natexis*239673
Rabobank Netherlands90016San Paolo Bank Holding153115BBVA*237747
BCI88594Generale Bank126889Caisses d’Epargne235660
NMB Postbank Group84194International Nederland Bank125343Dexia*232601
Cariplo82103Norddeutsche Landesbank118507Bankgesellshchaft Berlin220646
Monte dei Paschi di Siena75694Banco Santander114174Sanpaolo IMI*185403
Credito Italiano75223Cariplo107788Banca Intesa*179258
CIC Group74725Sudwest LB107602Unicredito Italiano*171730
BBV69986Crédit Communal de Belgique99941Nordeutsche Land Giro170759
Generale Bank67637BBV99174KBC Bank*163125
Nordeutsche Landesbank67515BNL98662Bank Austria*140161
Banques Populaires64701CIC97839BCI132188
Banque Indosuez55316Banca di Roma93373Banca di Roma122145
BBL50548BCI92449Merita Nordbanken112049
$ millions$ millions% increase$ millions% increase
Average top five21023731863151.5657698681.08
Average top three22806134143949.7164167287.93
Source: The Banker.

Banking groups that are the result of recent mergers.

Source: The Banker.

Banking groups that are the result of recent mergers.

12. The degree of concentration at the top is particularly striking in the smaller euro-area countries, where now just a handful of banks dominate these banking sectors. EU-11 countries’ banking systems are characterized by relatively few large banks, some of which are considered global players, and an array of medium-sized and small institutions. In almost all of the smaller countries, the top five banks hold more than 50 percent of the banking system whether measured by total assets, total loans, or total deposits (Table 9). In a few countries, the concentration is now even more pronounced. For example, in the Netherlands and Belgium, two large banking groups have more than half of banking sector assets, respectively. The four biggest countries have less concentrated banking sectors, although in France, the top five banking groups take in nearly 90 percent of all deposits. Notably, Germany has the lowest level of concentration in the euro area almost regardless of how it is measured. France and Spain are relatively more concentrated.

Table 9.Concentration Indicators in 1998 1/(In percent of total)
Sources: IBCA; and national central banks.

C5: share of the five largest banking groups. C2: share of the two largest banking groups.

Sources: IBCA; and national central banks.

C5: share of the five largest banking groups. C2: share of the two largest banking groups.

13. Mergers among smaller banks and savings institutions, though perhaps less visible, have been continuing for a number of years. For instance, in Germany, most of the recent merger activity has been among savings banks (Sparkassen), whose number has declined by some 5 percent, and credit cooperatives, dropping 13 percent over the last five years. In Italy, banking law changes in the early 1990’s have brought about consolidation in the segments of the banking system that were previously partly owned by the public sector (see subsection B below). Spain, too, has seen slow, but steady, consolidation in its savings bank sector, the cajas. For France, the consolidation of the mutual, cooperative and savings banks occurred earlier, but has proceeded further, with almost all of these institutions becoming a member of one of five large groups.8 What is notable about these mergers is that not only are they within-border mergers, but they also occurred within their own type. Some of the “unevenness” in the consolidation process, therefore, comes from the fact that, in contrast to the United States or United Kingdom, many of the EU-11 institutions are not available for acquisition outside their type owing to their unique ownership structures.

14. Because negotiating mergers and acquisitions (M&A), particularly cross-border ones, can sometimes become problematic, cross-border alliances have started to appear. For example, BSCH (Spain) has share exchange agreements with several European banks (Table 7). However, there are conflicting opinions as to whether all these alliances will eventually lead to fully-fledged cross-border merger operations. To date, there are only a few examples of large, cross-border M&A deals, which have a predominantly regional nature:

Dexia,9 Fortis, and ING in the Benelux countries; and Merita-Nordbanken-Unidanmark in the Nordic region. HSBC’s (Hong Kong/United Kingdom) proposed friendly bid for Credit Commercial de France (CCF) is pending as the first nonregional bank merger.

Share of foreign banks

15. Although individual euro-area countries’ banking systems are becoming more concentrated through mergers of large institutions, the concentration remains in domestic hands. Foreign bank shares within domestic markets are low within the four largest euro-area economies (Table 10). Of the four, market shares of foreign branches and subsidiaries as a share of total domestic assets range from around 12 percent in Spain to about 4 percent for Germany. At the opposite extreme is Luxembourg, where foreign banks dominate the banking scene: foreign branches and subsidiaries hold almost 100 percent of the domestic asset base. Ireland, too, has a large proportion of foreign branches or majority foreign-owned banks in its banking sector—over 50 percent. For the most part, the foreign institutions present in these international banking centers have been encouraged to locate there by various tax and other benefits and, for the most part, they do not participate in domestic credit creation, acting mostly as wholesale banks or locations to book assets in a favorable business environment.10

Table 10.Share of Foreign Banks 1/
From EEA CountriesFrom Third CountriesTotal
France 2/
Euro area
weighted average3.41.612.7
Source: European Central Bank, “Possible Effects of EMU on the EU Banking Systems in the Medium to Long Term”, February 1999.

Market share of branches and subsidiaries of foreign credit institutions as a percentage of the total assets of domestic credit institutions, end-1997.

1996 figures.

Source: European Central Bank, “Possible Effects of EMU on the EU Banking Systems in the Medium to Long Term”, February 1999.

Market share of branches and subsidiaries of foreign credit institutions as a percentage of the total assets of domestic credit institutions, end-1997.

1996 figures.

B. History and Industrial Policy Shaped the Current Structure

16. The present structure of the euro-area banking system is the outcome of historical development beginning in the nineteenth century. As in many banking systems around the world, a long history of government intervention and regulation of the banking system has left traces that are still pervasive today. In the euro area, two main features of its history can be highlighted: (1) policies directed at providing universal access to savings instruments, and the provision of credit to targeted segments of the economy, often accompanied by special privileges granted to specific institutions; and (2) the nationalization and subsequent privatization of large portions of national banking sectors.

Unique role of savings institutions

17. Although there is a rich history of banking in Europe, the development of present-day savings institutions is normally attributed to policies that attempted to encourage the provision of financial services to sectors or parts of society that had been left outside the main channels of economic development. At the turn of the nineteenth century, savings banks were envisioned as instruments for the channeling of households’ savings to the industrial sector, whereas mutual and cooperative banks were the providers of credit and liquidity management services for nonindustrial sectors of activity, such as artisans or farmers.

18. The regulation and organizational structure of these institutions vary across countries, but most of them share the feature that their first priority is not to make profits but to accomplish some well-specified public goal. In France and the Netherlands, savings and/or mutual banks are concentrated in a few groups with a pyramidal structure.11 In Germany and Italy, by contrast, the savings and cooperative sector consists of hundreds of small independent entities, which have contributed to the highly diverse nature of their banking sectors. Spain lies in between in terms of concentration, as the cajas have undergone a steady process of consolidation among themselves, mainly at the regional level. In all of these cases, special ownership structures provide these institutions with the opportunity to pursue other goals besides generating profits. Though profitability is gaining increased importance even to this class of institutions, to the extent that they choose not to maximize profits, this gives them a competitive edge relative to commercial banks, which feel the pressures of their shareholders.

Public ownership

19. The heavy weight of the public sector at some point in history is another common characteristic of Europe’s banking sectors. This feature has decisively shaped the banking market, either through the subsequent privatization or through the influence exerted by the public institutions on the types of products and their prices (Box 2). Although most EU-11 central governments no longer own or control large segments of the banking sector, in some countries, notably Germany and Austria, state and local municipalities continue to do so. The Landesbanken and Sparkassen capture over one-third of total assets in Germany, and similar institutions hold about 14 percent of total assets in Austria.

Special privileges

20. An important auxiliary component of public policy in the banking sector has been the granting of special privileges to some of these institutions to facilitate the fulfillment of their role in directing credit to particular areas and in providing savings vehicles for some segments of the population. These privileges have taken several forms.12 In France, the savings banks enjoy the monopoly distribution of the most popular tax-exempted saving account, the Livret A, and some of the mutual networks had, until recently, the monopoly for the distribution of subsidized loans to some sectors of activity. In Germany, the Landesbanken receive two formal government support mechanisms that influence their credit rating and help to lower their funding costs. Implicit guarantees are also present in other countries, as evidenced by lower bank ratings when banks are considered on a “stand alone” basis (i.e. without such guarantees or implicit government support), such as in Moody’s financial strength ratings (Table 11).

Table 11.Banks’ Financial Strength Ratings(Proportion of banks in each rating)
Number of rated banks29352615
Source: Moody’s.
Source: Moody’s.

Box 2.Privatization of Financial Institutions in Major European Countries

Privatization has been uneven in Europe. The privatization of banks in major European countries started in the mid-1980s (e.g., with the sale of Société Générale), but acquired momentum only in the 1990s, with the sale of banks such as BNP and Crédit Lyonnais in France, Argentaria in Spain, and several major banks in Italy. As well, many Italian savings banks changed their ownership structure from exclusively “foundation-owned” to more open, shareholder-based, entities.

The transformation of the Italian banking sector has in some ways proceeded the farthest. Public sector ownership of the banking sector in Italy fell from 72 percent of total assets in 1993 to 16 percent in 1999, when the share controlled by the Treasury represented less than 1 percent of banks’ capital. The share of “foundations” has also declined substantially, with their stake in large banks being cut by two-thirds and in savings banks by half. Despite this impressive progress, the government continues to encourage foundations to divest and relinquish their control over banks through a set of fiscal incentives and sanctions.

Privatization in France has also been significant, although it leaves room for entrenchment by the public sector. On the one hand, the privatization of the three major commercial banks resulted in a ownership structure with significant participation of international institutional investors and banks that has imparted some market discipline to the sector. On the other hand, several of the financial institutions sold by the state in the late 1990s were acquired by mutual banks (CIC by Crédit Mutuel) or publicly controlled institutions (Crédit Foncier by the Caisses d’Epargne). This outcome was attributable in large part to the stronger capitalization of mutual banks and to their more accommodative approach to labor redundancies, which helped the government disregard foreign bids. In addition, the recent reform of the Caisses d’Epargne fell short of fully privatizing their operations: they were mutualized, and are now under the control of depositors, local governments, and the state-owned Caisse de Dépôts et Consignations (CDC).

Change in Germany has been slow and limited. This stability has in part been attributed to the demands created by German Unification, which renewed the mission of public banks to promote regional development, as well as to a conservative German populace with respect to their local savings banks. Recent decisions by the European Commission may, however, help catalyze changes, at least for the Landesbanken who will likely be required to pay market rates for their funding.

Finally, the share of the public sector in Spanish banking has increased, rather than decreased, in the last 15 years. This has not been the outcome of nationalization: the Spanish authorities quickly auctioned off the banks in which they had intervened during the financial crisis of 1978-83, sold Banesto within ten months of its intervention in 1992, and sold off, under the umbrella of Argentaria the traditionally state-owned specialized banks. Rather, the increasing importance of the public sector is the result of the growing weight of savings banks. Although technically private institutions, local and regional government officials sit on the governing boards of the cajas and influence policy. Although they were not considered as part of the Spanish banking system until the 1970s, they were subsequently allowed to carry out universal banking activities (1977) and expand their operations beyond their original regions (1988). Today, some of them have become major players: the two largest, Caja Madrid and La Caixa, have nation-wide operations and rank among the top five Spanish banks.

21. There has been growing recognition that the private sector may be better situated to meet regional or local development goals and provide universally-available savings vehicles. In the past, some governments believed that credit would not be allocated to satisfy various public development goals, particularly following World War II, and that small investors would not have access to savings vehicles necessary for their retirement. Most euro-area governments now take the view that, while directed lending and savings incentives may still play a role in some selected circumstances, most of the intermediation process should be free from government intervention.

C. Recent Performance and Changes in Balance Sheet Structure

22. Despite the persistence of relatively high labor costs and fierce competition for traditional banking business, the performance of EU-11 banks improved in the second half of the 1990s. Financial liberalization, globalization, and technological changes over the past years have challenged banks to strengthen their competitiveness, and many of them have reacted proactively to the new environment. Facing strong competitive pressures on their traditional income-generating activities, many banks reacted by diversifying their income sources both across products and countries. The structure of banks’ balance sheets reflects these changes: on the liabilities side, traditional deposits have shrunk to the benefit of money market mutual funds and other liabilities, while on the assets side, banks have developed trading activities and securitization operations. Banks’ capitalization has also shown signs of improvement and—in those countries where this important, but sensitive, statistic is disclosed—the share of bad loans in total assets has declined, often following the strengthening of the economy. On balance, despite high operating costs, profitability measures of the banking sector generally improved. Performance, however, differed across types of banking institutions. In France and Spain, savings and cooperative banks increased their share of the sector’s profits, while their share declined in Germany and Italy. This section reviews these developments and highlights some trends that have emerged in the region in the business of banking.

23. The data for this section come predominantly from annual reports of individual banks reporting to FitchIBCA, which harmonizes bank accounting definitions across countries. Although the set of banks covered is not comprehensive, use of the FitchIBCA data base has two main advantages.13First, FitchIBCA uses the domestic accounting-based source data and adds or subtracts the appropriate categories to provide cross-country consistency (termed “global” variables). This makes comparisons possible—for instance, national sources provide capital ratios that are not comparable across countries because definitions of capital differ across countries. Second, the FitchIBCA database contains many more variables that have been made comparable than other sources (there is only a limited number of them in, for example, the OECD Bank Profitability statistics). This makes it possible to extend the analysis to cover other information such as off-balance sheet items.

Expanding business focus

24. Deregulation of the financial services industry in the euro area over the last 15 years has considerably increased competition in the banking sector and reduced the role of traditional intermediation activities as a source of income for banks. Between 1992 and 1998, net interest margins (NIM)—the difference between banks’ revenues from lending and the remuneration of deposits—have declined from 2 to 1.5 percent of banks’ assets (Table 12). Even though net interest margins have also declined in the United States, they are now lower in all euro-area countries (with the exception of Italy, Spain and Portugal) than in the United States. The steepest drops were recorded in small countries such as Ireland and Portugal, where margins were quite high to start with and for which the path toward EMU accession was associated with falling inflation and domestic interest rates. A similar, albeit more moderate, process also occurred in Italy.

Table 12.Performance Indicators for Banking Systems
Source of revenue
Net interest margin706653757163787158787467736959575556495248706157
Net commission inc.172127161211171527182026161722283029556774
Net financial income121218041027835567117761814127710
Net fee income000370000012556669
Net trading income1212180132783556697761396111
NIM to average assets1.
Costs to income 1/94898577777881847789807585817886616965656584127159
Of which:
Personnel costs5542574954496158555941575653364743464645292720
Earning power1.
Source of revenue
Net interest margin736356747361636060707163475943716762776457
Net commission inc.171821810162224222222233243119212691016
Net financial income067121219769425713867710109
Net fee income00000000000000
Net trading income067121219769425713867710109
NIM to average assets1.
Costs to income 1/87858585806513198647263575753.51807883848379
Of which:
Personnel costs595655615954374646575955472447573855563344
Earning power0.0-
Sources: FitchIBCA Inc; and Fund staff calculations.Notes: Performance indicators are defined as follows: NIM denotes net interest margins; ROA, or return on assets, is the ratio of pre-tax profits to average assets; ROE, or return on equity, is the ratio of pre-tax profits to average equity; earning power is the ratio of pre-provisioning profits to average assets. The first year reported for the USA is 1993 since the sample of banks reporting in the database is not representative.

Ratio of operating costs to operating income. Our definition of operating costs differs from the traditional one in mat it excludes provisions, which are too year- and bank-specific.

Sources: FitchIBCA Inc; and Fund staff calculations.Notes: Performance indicators are defined as follows: NIM denotes net interest margins; ROA, or return on assets, is the ratio of pre-tax profits to average assets; ROE, or return on equity, is the ratio of pre-tax profits to average equity; earning power is the ratio of pre-provisioning profits to average assets. The first year reported for the USA is 1993 since the sample of banks reporting in the database is not representative.

Ratio of operating costs to operating income. Our definition of operating costs differs from the traditional one in mat it excludes provisions, which are too year- and bank-specific.

25. Other sources of income, notably commissions from asset management and other services, have come to represent a significant share of banks’ revenues. The new environment created strong incentives for banks to look for new sources of income. On the liabilities side, they began diversifying into businesses with which synergies could be found, such as the management of investment and pension funds. In several EU countries, most of the institutional investors are now included in banking groups and operate with the same corporate strategy. As a result, one-fourth of banks’ operating income in three of the four largest euro area countries came from commissions remunerating this activity in 1998, up from a sixth in 1992. In a significant banking center such as Luxembourg, asset management now provides almost a third of banks’ income.

26. The expansion of commission income has been principally based on the impressive growth of mutual funds, a line of business in which banks in most euro area countries have secured a prominent role. Among the four largest countries, Italy and Spain experienced the strongest growth in managed assets in the second half of the 1990s. The stock of assets held by mutual funds in Italy has multiplied by five since 1995, while it has more than doubled in Spain. The growth of mutual funds in France in recent years has not been as strong as elsewhere, but the French mutual fund industry is mature relative to other European countries: with more than euro 600 billion in assets, it is the biggest in Europe (Table 13). By contrast, mutual funds play a secondary role as savings vehicle in Germany, as time and savings deposits are still the preferred financial instrument of most individual investors.14 The growth of mutual funds has benefited banks in France, Italy and Spain because banks in these countries have been in the forefront of the industry. Indeed, assets held by mutual funds affiliated with banks account for more than two-thirds of the total assets of the industry.15 Many large banks have also taken advantage of their distribution networks—as well as the asset management capabilities they had developed in support to mutual funds—to carve out a comfortable position in the insurance market as well.

Table 13.Managed Savings in Europe(1999Q3, in billions of euros)
Luxembourg440United Kingdom305
Source: European Federation of Investment Funds and Companies (FEFSI).
Source: European Federation of Investment Funds and Companies (FEFSI).

27. Of course, the counterpart to the growth in managed savings has been the decline in customers’ deposits in banks’ balance sheets, and banks have developed other sources of funding. The expansion of mutual funds has been associated with a decline in the share of deposits in banks’ liabilities, by one-sixth in Ireland and around 10 percent in four other EU-11 countries between 1995 and 1998 (Table 14). Banks have however partially offset this decline in one of their sources of funding by channeling some of the short-term mutual fund investments toward the purchase of certificates of deposit (CDs) or banks’ debentures—a practice relatively common in France for many years.

Table 14.Main Components of Banks’ Balance Sheets, 1998(In billions of U.S. dollars and percentage of total assets)

Total assets4,7116,8062,0061,3374779302732565041,3983181,7292,121
Total deposits3,1944,3841,1561,1103308141611934169572601,1801,367
Total loans—net1,9203,4781,0846772623491601451058101608321,036
(Percentage of total assets)
Total loans—net40.851.154.050.654.937.558.556.620.957.950.448.511.3
Interbank assets27.423.717.014.425.422.512.115.848.012.522.421.910.1
Other earning assets23.121.619.525.314.834.219.420.327.324.318.122.55.2
Total non-earning assets7.
Customer deposits39.135.333.956.435.359.144.243.436.345.755.144.09.2
Interbank deposits28.828.123.024.633.928.514.931.146.322.826.728.17.9
Money market funds8.
Other funds and liabilities18.830.710.48.926.S8.917.47.77.311.85.914.18.4
Equity and reserves4.
Total contingent liabilities30.414.431.05.311.1105.717.
No. of banks in the sample3431583502151767512431205242
Structure in 1995
Customers deposits38.239.933.160.436.962.443.852.540.451.361.147.310.7
Total contingent liabilities23.713.125.76.613.
No. of banks in the sample356170351911173769321165140
Structure in 1992
Customers deposits33.740.130.554.547.665.621.964.621.753.464.345.316.7
Total contingent liabilities23.713.125.76.613.
No. of banks in the sample272527130852331611793432
Percentage change 1995-98
Customers deposits/assets2.2-11.52.5-6.6-4.3-5.31.0-17.3-10.3-10.9-9.9
Contingent liabilities/assets28.39.820.4-19.5-14.428.8-1.29.552.2-45.010.9
Sources: FitchIBCA Inc; and Fund staff calculations.
Sources: FitchIBCA Inc; and Fund staff calculations.

28. To sustain their profitability, banks have also developed their presence on the stock market, and income from trading securities on banks’ own account has become an important, albeit uneven, source of revenue. On the asset side, banks have diversified their sources of income by offering their services for security trading and underwriting for the accounts of their customers. Anecdotal evidence also suggests that, with the advent of EMU, fee income from the issuance of securities by euro area companies and the income earned on advising mergers and acquisitions have been a fast-growing source of revenue for banks. In 1998, securities trading operations provided about one-fifth of banks’ operating income in France, increasing from the level in the beginning of the 1990s. In other EMU countries, however, this type of activity even if expanding contributes relatively little, and generally less than in the United States, to banks’ operating income. Of course these revenues vary substantially from year to year, reflecting their sensitivity to market developments.

29. Despite a disintermediation of finance, EU banks’ asset structure reflects the rapid increase of lending since the run-up to, and advent of, EMU. Perhaps surprisingly, the share of loans in banks’ balance sheet assets has increased in some countries, even though lending spreads have generally declined. This process started before the introduction of the euro, reflecting growing demand for credit owing to the downward path of interest rate levels and an increase in the degree of competition in the bank loan market. Mortgage lending has grown substantially in a number of euro area countries and the process of securitizing these loans, while still at an early stage in some countries, is also progressing at a brisk clip. After the euro was introduced, lending growth was amplified by the growth in “leveraged” loans (e.g., to finance mergers and acquisitions as European corporations restructured), which overshadowed the also substantial increase in issuance of corporate bonds.16 More generally, the increase in the share of loans in banks’ balance sheet may reflect the slow take-off of securitization of standard loans, as opposed to that of mortgages, owing partially to remaining ambiguities in some countries’ legal framework covering such instruments. However, demand factors, such as the growth of institutional investors in the euro area, and supply factors—such as the desire of smaller banks to diversify their assets and the generally favorable regulatory treatment of securitized assets—are however likely to sharply increase the use of loan securitization in coming years.17 Lastly, the decline in non-loan assets represented by the reduction in government securities outstanding in the run-up to the euro, also acts to increase loan shares.

30. The growth of bank lending also results from banks’ expansion into foreign markets, as evidenced by the substantial rise in international claims that they hold. The ratio of foreign assets to domestic private sector claims in banks’ assets increased in almost all EU-11 countries in 1995-98 (Table 15), in part prompted by the low interest rates prevailing in Europe in the period. In 1999, this ratio was larger than 1 in three countries (Belgium, Luxembourg, and Ireland), and above 0.75 in another four countries. It declined only in Portugal and Spain, owing to the unusual increase in domestic lending that accompanied the strong economic performance in these two countries.18 Expansion into emerging markets has tended to reflect old economic links. Foreign exposures of Spanish and Portuguese banks, for instance, have been concentrated in Latin American countries, while German and Austrian banks have accumulated significant claims against eastern Europe.

Table 15.Foreign Asset Exposures
Panel A: Foreign Assets as a Percentage of Claims to the Domestic Private Sector (CDPS)

1998 1/
Change in ratio

(in percent)
Change in CDPS

(in percent)
Panel B: Consolidated International Claims on Individual Countries (June 1999)

Total (US$ billion)6601,313222158992812527417356412
Percentage in total claims of claims against:
Other EU-11 countries3737403535472522383534
Non EU-11 countries6363606565537578626566
“Reporting” countries 2/3640352430415868427676
Eastern Europe14319122233
Offshore centers875610545666
Developing countries13810281038191010
Of which:
Latin America337262111456
1995-1999 percentage growth in nominal terms of claims against:
Non “Reporting”

Eastern Europe6511-1755131401251626955
Offshore bank centers14-19-46757-24-62188551914
Developing countries2245925011-14207104587787
Of which:
Latin America24394633310354401035108130
Sources: IFS (Panel A); BIS; and Fund staff calculations (Panel B).

Data for France, Germany, Belgium, and Luxembourg refer to 1997.

“Reporting” countries are the EU-11 excluding Portugal, plus Canada, Denmark, Japan, Norway, Sweden, the United Kingdom and the United States.

Sources: IFS (Panel A); BIS; and Fund staff calculations (Panel B).

Data for France, Germany, Belgium, and Luxembourg refer to 1997.

“Reporting” countries are the EU-11 excluding Portugal, plus Canada, Denmark, Japan, Norway, Sweden, the United Kingdom and the United States.

Operating costs

31. Operating costs, despite some decrease, remain high and absorb at least three-fourths of banks’ operating income in Germany, France, Italy and Spain. The development of non-interest-earning activities does not seem to have resulted in a significant reduction of banks’ operating costs. The development of asset management activities, for example, requires increased fees to be paid to financial advisors and the development of some fee-generating activities may entail higher information technology (IT) and personnel costs. Indeed, although the ratio of labor costs to operating income declined, to a different extent in almost all euro-area countries, it remains higher than in the United States (except in Luxembourg, Finland, and Portugal). Personnel costs remains the largest component of operating costs, with the average euro-area bank spending over half of its operating expenseson wages and salaries in 1998, despite the rise in expenditures associated with the conversion to the euro.

Asset quality

32. The assessment of banks’ asset quality is made difficult by the lack of uniformity in the treatment of nonperforming loans across euro-area countries, but it appears to indicate an improvement in most countries, notably those more advanced in the economic cycle (Box 3). Excluding Germany, Austria, Luxembourg, and the Netherlands that do not publish figures on the level of nonperforming loans (NPL), EU-11 countries can be divided in two groups with respect to the share of bad loans in total loans (Table 16).19 In the first group, including Finland, Ireland, Portugal, and Spain, NPL account for less than 3 percent of total loans. In the second, accounting for Belgium France, and Italy, NPL range from 3 percent to 9 percent of total loans. Remarkably, the share of bad loans in Finland, Portugal, and Spain has dropped sharply to between 2 percent and 3 percent since the mid-1990s, in contrast to a more subdued reduction in France and a very modest decline in Italy.20 The pace at which bad loans have been written off in different countries has been related to the stage of these countries in the economic cycle, but it has also reflected individual country policies. Those countries in which supervisors can have a say on the amount of provisions that are tax deductible (e.g., Portugal and Spain) have typically witnessed a faster pace of write-offs than those countries in which deductibility is more inflexible (e.g., Italy).

Table 16.Measures of Liquidity and Solvency(In percent; 1998 unless otherwise marked)

Loans/customer and ST funding53.277.981.961.079.342.975.366.525.379.360.363.917.9
Liquid assets/customer and ST funding21.033.038.833.325.621.219.328.739.632.029.27.3
Asset Quality
Non-performing loans/total loans
Provisions/NPL (1998)
BIS ratio 5 largest banks11.010.19.612.012.711.011.912.513.014.011.411.71.3
BIS ratio (authorities)n.a.n.a.n.a.n.a.n.a.n.a.11.515.7n.a.10.8n.a.12.72.7
Equity/total assets4.
Equity/net loans
Change 1992-1998, in percent32.1-10.9-6.4-0.8-15.135.5-3.423.332.44.7-33.72.8
Subordinated debt/total assets1.
Number of Banks in the Sample
Sources: FitchIBCA Inc; National Central Banks; and The Banker.Note: The term “n.a.” implies an absence of an aggregate representing the banking sector as a whole. The number of banks reporting these data to FitchIBCA voluntarily is small and thus these data have not been reported as their representativeness is not certain.
Sources: FitchIBCA Inc; National Central Banks; and The Banker.Note: The term “n.a.” implies an absence of an aggregate representing the banking sector as a whole. The number of banks reporting these data to FitchIBCA voluntarily is small and thus these data have not been reported as their representativeness is not certain.


33. The development of non-interest income activities may have had a positive effect on banks’ performance, but profitability remains generally lower than in the United Kingdom and the United States. Non-interest income has been the most dynamic component in the bank income structure in recent years, but the high costs associated with these activities seem to have impeded the full benefit of this evolution on banks’ profitability. The improvement of profitability of euro-area banks over the past decade, measured in terms of returns on assets (ROA, the ratio of pretax profits to assets) and on equity (ROE, or pretax profits to equity), is probably a result of the greater attention that commercial banks have paid to profitability, as well as a consequence of the region’s economic recovery since the 1992-93 slump21. Indeed, in countries such as Germany and Austria—where the contribution of non-interest income remains small, economic recovery has lagged and foreign exposure has required important provisions for bad loans: profitability measured by ROA actually declined. By contrast, in countries such as Ireland, Portugal, and Spain, profitability has been relatively high—more in line with that in the Anglo-Saxon countries. In these euro-area countries and Italy, banks’ ability to generate a profit from their current operations (as measured by their “earning power,” the surplus of operating income over personnel and administrative costs relative to assets) was in 1998 close to U.S. levels.

Box 3.Disclosure of Asset Quality and Bank Strength

Cross-country comparisons of asset quality should proceed cautiously because classification schemes for problem loans and public disclosure policies in EU-11 countries vary widely. Compared with the rules governing loan classification in the United States (and now Japan), rules in many EU-11 countries are not explicit, allowing banks a degree of discretion. Several countries (e.g., Germany and Austria) do not have explicit rules that classify loans as nonperforming loans after they have not been serviced for a certain number of months, leaving ample room for discretion on the part of banks. The merit of such approaches, which are generally rationalized by the intimate knowledge that banks are deemed to have of their debtors, is less clear when banks expand outside domestic markets and into markets where the home supervisors and the banks themselves may not have a long track record. In some other countries, rules for the automatic classification of troubled loans do exist, but are relatively generous. In Italy, for instance, until recently loans that were not being serviced might have been classified as doubtful (partite incagliate) for periods of up to 18 months before being recognized as bad loans (sofferenze) (Bank of Italy, “Relazione Annuale—appendice,” several years).

Even if rules for classifications are clear, some EU-11 supervisors believe that the public disclosure of figures on nonperforming loans (NPL) could be misinterpreted by markets and depositors. In addition, supervisors also point to legal impediments or jurisdictional issues that impede disclosure. In Germany, for instance, mandatory public disclosure would require a change in the Commercial Code.

Credit rating agencies and others attempt to supplement the information disclosed by banks, and assess banks’ strength even in the absence of relevant data. For instance, free capital, usually measured as capital minus fixed assets, participations, and bad loans—can provide a more accurate picture of banks’ strength than crude measures such as equity-to-assets ratios, since it measures the amount of capital readily available to cushion new losses. Unfortunately, lack of consistent NPL data renders the computation of such a measure difficult. Barring this more detailed computation, credit ratings can still provide a proxy where the paucity of detailed data hampers an independent evaluation. Moody’s and other rating agencies have increasingly rated European banks from different perspectives, with an emphasis on a stand-alone basis, using a methodology which ignores implicit or explicit government support (Table 11). The resulting distribution of ratings may not be an unbiased measure of the strength of distinct banking systems, however, because of a self-selection problem arising from the fact that ratings occur at the request of banks. But with appropriate caveats, it is informative.

Credit ratings provide a differentiated cross-country picture, in which the weight of different types of banks play a role. Ratings suggest that the banking sector in Spain is the most creditworthy within the four largest EU-11 countries, a classification that meshes with the strong capitalization of both private and public banks in the country. Interestingly, the Italian banking system fares better with respect to ratings than the French system. This is likely to reflect in part the fact that mutual banks, which are well-capitalized in both France and Italy, are more integrated in France (e.g., under Crédit Agricole) than in Italy, leading to a larger count of strong banks in the latter country. Germany, although having lost the high ratings that once characterized its top banks, have nonetheless been able to avoid a slip toward ratings below D.

Bank capitalization

34. Challenged to adapt to the new environment, most EU-11 banks have attempted to boost their capitalization, a key condition to expand their business and minimize the cost of funds raised in the financial markets. The Capital Adequacy Directive and the Basel Committee’s recommendations on bank soundness now require banks to also hold capital to cover against market risks (in addition to credit risks) and banks wanting to expand into wholesale banking, for instance, have had to increase capitalization. The BIS risk-weighted capital to asset ratio of the 5 largest banks in 10 of the 11 euro-area countries now exceeds 10 percent (it stands just below 10 percent in Italy). Greater capitalization also helps reduce the cost of borrowed funds for banks, at a time when the fall in the share of savings held in deposits has made banks increasingly reliant on money and capital markets. To improve capitalization, banks have put a premium on raising equity and quasi-equity.22 Subordinated debt remains, however, a popular means for raising capital, because it offers slightly higher rates—in the otherwise low interest rate environment prevailing since the mid-1990s—in exchange for lower seniority in cases of default. Still, despite the development of such common patterns across countries, capitalization, measured in terms of the equity-to-asset ratio, varies considerably in the euro area: it ranges from 3.4 percent in Germany to 6.4 percent in Spain, with Italy in the middle. Idiosyncratic factors, such as the presence of so-called “hidden reserves” (e.g., in Germany), partly explain these differences.23

Performance by institution type

35. Banks’ revenue sources and performance have differed significantly across types of institution. Net interest margins have declined less in savings and cooperatives than in commercial banks. The particular involvement of the savings and cooperatives banks in the provision of credit to individuals and to small and medium-sized enterprises at the regional level implies that these banks rely more heavily on interest income than commercial banks. Despite the competitive environment in traditional lending, net interest margins have decreased less as a percentage of assets for savings and cooperatives than for commercial banks (Table 17). Such different evolutions can in part be explained by the comparative advantage that savings and cooperatives may have in funding themselves at lower interest costs. In particular, in some countries these institutions may benefit from implicit or explicit government guarantees, lowering the risk of public or quasi-public institutions and thereby lowering funding costs.24 In addition, many of these institutions face a relatively inelastic supply of deposits—their depositors are quite loyal and tend to leave their deposits with them even when deposit rates decline relative to other investments.

Table 17.Performance Indicators by Type of Banking Institution in Selected Euro-Area Countries(In percent)
Commercial BanksSavings BanksCooperatives
Net interest margins/assets2.
Net profits/assets (ROA)
Operating costs/income93.391.485.
Share of sector’s profits6435332938
Commercial BanksSavings BanksCooperatives
Net interest margins/assets2.
Net profits/assets (ROA)
Operating costs/income101.184.066.368.073.875.3
Share of sector’s profits25332314189
Commercial BanksSavings BanksCooperatives
Net interest margins/assets2.
Net profits/assets (ROA)
Operating costs/income82.077.871.576.376.274.5
Share of sector’s profits344533141818
Commercial BanksSavings BanksCooperatives
Net interest margins/assets3.
Net profits/assets (ROA)10.90.811.41.3
Operating costs/income78.575.877.069.166.865.0
Share of sector’s profits6551173112
Sources: FitchIBCA; Bundesbank Monthly Reports (various issues); and Fund staff calculations.
Sources: FitchIBCA; Bundesbank Monthly Reports (various issues); and Fund staff calculations.

36. Comparisons of profitability measures across bank types are not straightforward because of the different objectives and constraints faced by each of them. Savings and cooperative banks are often “nonprofit” institutions and may not face pressure from shareholders to generate profits or distribute dividends, though in recent years the focus on profits as a gauge for the efficient transfer of resources has become more prominent. On the other hand, most of these institutions cannot issue marketable equity, and thus can grow only by accumulating profits. Moreover, some of them benefit from a favorable tax treatment, which increases after-tax profits relative to banks without such advantages.

37. Thanks partly to a strong regional presence and the specificity of the products they offer, the profitability of cooperative and savings banks in some countries appears strong.Table 17 suggests that the profitability of savings and cooperative banks is similar to that of commercial banks in France and Germany, and is somewhat larger in Italy and Spain. The performance of cajas and cooperatives is strong in Spain, and the cajas, with a firm hold on the provision of domestic credit and financial services, have been gaining market share in recent years. In Italy, savings banks, which are prominent in the wealthy Northern provinces, have also been able to post figures suggesting good performance vis-à-vis commercial banks, several of which still carry the burden of their past as public banks. Note that market share dynamics are harder to evaluate in Italy, where institutions classified as savings banks in the past now have a status of commercial banks since their reform, in the beginning of the 1990s. In France, the heritage of some of the leading commercial banks (e.g., Crédit Lyonnais) and high labor costs weigh on the profitability of commercial banks, while the guaranteed income from passbooks (remunerated according to a flat fee by the government) and their dominant positions in certain regions has helped to boost the profitability of savings and some mutual banks.

38. The various types of banking institutions in EU-11 countries differ by their levels of capitalization (Table 18). In part owing to restrictions on the distribution of profits, savings, cooperative, and mutual banks generally exhibit a ratio of equity to total assets significantly higher than that of commercial banks. Although this difference is somewhat less dramatic for France and Germany, it exceeds 3 percentage points in Italy and Spain.

Table 18.Comparative Structure of Balance Sheet by Type of Bank (1998)
Total Assets
(In billions of U.S. Dollars)
(Percent of total assets)
Net loans
Customer deposits
Interbank deposits
Contingent liabilities (credit equivalent)
Sources: FitchIBCA, Inc.; Bundesbank, Bankenstatistik, February 1999 for German Savings and Cooperatives.
Sources: FitchIBCA, Inc.; Bundesbank, Bankenstatistik, February 1999 for German Savings and Cooperatives.

39. The balance sheets of savings and cooperative banks show some regularities resulting from their specific business specialization and client bases: traditional lending and deposit-taking activities still represent the cornerstone of their activity and contingent liabilities still weigh little in their balance sheets. Two features seem to emerge. First, on the assets side, the importance of loans varies by type of bank, and it does so differently across countries. The share of loans in total assets is low for savings banks in France, but high for savings banks in Spain, because the funds collected by French savings banks through passbooks are largely redirected to other public institutions, while the cajas in Spain play a key role in direct lending at the regional level. The disparity between the share of loans in the balance sheets of savings banks and commercial banks in Spain is also exacerbated by the large equity holdings of Spanish commercial banks. Second, on the liabilities side, the weight of customer deposits is typically heavier for savings banks than for commercial banks, which tend to rely to a greater extent on interbank deposits. This difference is most notable in the case of France, where passbooks (Livrets) are among the most important products offered by savings banks, but this feature is also present in other countries and, in various degrees, in cooperative banks.

D. The Rapid Expansion of Capital Markets Since Monetary Union

40. The introduction of a common currency and a common monetary policy has served to quicken the pace of capital market activities, to which banks must now respond. As previous sections have emphasized, the institutional structure of the euro-area banking sector has been undergoing a metamorphosis for some time. The process of disintermediation, and euro-area universal banks’ reaction to it, have been reflected both in their income statements and in their balance sheets. However, recent capital market developments are forcing banks to adapt ever more quickly. A few of these new developments are outlined below.

41. The first year of monetary union was characterized by a sharp increase in the volume of publicly traded debt instruments issued by the private sector, a substitute for intermediated finance. Private sector security issuance surpassed all expectations, despite the remaining cross-border limitations imposed on some markets. At the short end of the maturity spectrum, the stock of certificates of deposit (CDs) issued by monetary financial institutions rose by 50 percent, while that of commercial paper rose by more than 40 percent. Thus, the market for privately issued short-term debt overtook that for government paper. The remaining fragmentation of short-term debt markets reflects to a great extent the existing distribution channels, which foster the allocation of the majority of these instruments to traditional domestic customers or asset funds associated with the issuing institution. In addition, cumbersome cross-boarder settlement procedures, a result of the still inadequate infrastructure of these markets, represent a relatively high barrier for short maturity paper. The large number of depositories, custodians, and settlement systems for sovereign and corporate debt has meant that these secondary markets do not yet act as an integrated whole, even when price quotes are made in a common currency. However, this problem is currently being addressed by the authorities and is expected to lessen over time.25

42. The boom in new corporate bond issues, rising 70 percent in 1999, revealed the already significant depth of primary bond markets and the existing ability to market new issuances across countries. This integration in part reflected the earlier efforts made by several large sovereign issuers to reach investors across borders in the run-up to EMU, as well as the disappearance of the currency risk. The relative ease with which these new issuances were absorbed also reflected the demand of increasingly prominent managed savings (mutual funds, life insurance, and pensions), as well as the favorable conditions provided by low short-term interest rates.

43. The share of lower grade (high-yield) bond issues also increased sharply, propelled by a number of large M&As. The issuance of private high-yield bonds tripled in 1999, albeit starting from a very low base. Although notable, this growth was overshadowed by the growth of other high-yield instruments (e.g., leveraged loans) and was held back somewhat by the insufficient harmonization of bankruptcy procedures in the region. The latter continued to render relatively difficult the homogenous pricing of assets with a sizable risk of default.

44. The expansion of these market instruments heralds a dramatic change in the financial landscape of the euro zone. While the role of bank loans is important and still increasing, the emergence of this new environment presents banks with the challenge to adapt to a much more capital market-based financial system.

E. Summary

45. Overall, the picture of the euro-area banking system arising from this analysis of the data suggests the following:

  • The financial system in the euro area remains bank-dominated, but disintermediation and capital market developments are changing the balance sheet structure of euro-area universal banks.
  • Different types of banks co-exist, though they are segmented by country. Savings and cooperative banks still hold large market shares in many euro-area countries.
  • The trend of consolidation has occurred predominantly within borders (without the inclusion of foreign banks) and has raised concentration levels.
  • Profitability has been maintained as a result of a diversification of revenue sources into non-interest-based revenues. Operating costs remain high.
  • The composition of liabilities is shifting from insured deposits to uninsured market-based funding.
  • Most banks are attempting to boost their capitalization in order to expand their activities and reduce the cost of funds raised in the financial market.

III. Possible Policy Implications

46. The structure of the European banking system and the recent trends evident in the data presented above pose several policy issues—the goals of consolidation and competition policy, the role of ownership and control structures, and supervisory and regulatory oversight. The following questions are thus explored.

  • Is financial disintermediation endangering the profitability of the euro-area universal banks or do the banks have the flexibility to respond to the decrease in traditional bank business? Will these new revenue sources continue to generate profits?
  • Consolidation has been rationalized on the basis of the economies of scale and scope it can generate, which can vary between the retail and wholesale lines of business. Are the conditions for realizing these benefits—free access to new markets and flexible use of inputs, including labor—fulfilled in the euro area?
  • Could the prevalence of consolidation at the national level rather than across borders, particularly at the retail level, hurt competition? To what extent could it counter the benefits of technological advances such as Internet banking?
  • A common feature of euro-area banking systems is the substantial share of financial institutions whose ownership structures mean that they are “not for sale.” In light of the questions addressed above, how could this feature affect the reshaping of the financial system in the region?
  • Taking into account that the “optimal size” of banks may have grown, do existing impediments to cross-border consolidation promote the creation of “too big to fail” institutions within each country? In view of this changing landscape, how is the financial stability and soundness of the banking system affected?
  • Are current supervisory structures appropriate for this rapidly changing environment, in particular as regards the development of large, complex institutions; the increased competition from potential entrants benefiting from new technologies; and the greater access of euro area banks to interbank credit following monetary union?

A. Is Financial Disintermediation Jeopardizing Euro-Area Universal Banks’ Profits?

47. Financial disintermediation may not translate into a major loss of business for the banking sector, but the recent growth in income fees from certain market segments may flatten in the near future. The deregulation of financial activities has opened new areas of activity for European banks since the mid-1980s. As noted in Section II, banks in several countries have taken up these opportunities, and this has translated notably into a rise in income from fees and commissions. Looking forward, greater reliance of firms on other forms of financing need not hurt banks, since traditional corporate lending commands very thin margins anyway (even in the United States, this margin averages 1.0-1.5 percent, and in some countries such as France it is sometimes negative). Issuance of publicly traded instruments, on the other hand, can generate fees for advising and underwriting services, as well as the provision of further guarantees and stand-by credit lines (e.g., to back up the issuance of commercial paper). However, optimism about the profitability of the sector should be somewhat moderated. Expectations of returns on equity around 20 percent could prove excessive.26 The recent growth in demand for mutual funds, for instance, is likely to abate somewhat, as it reflected in part a portfolio shift away from simple holding of public debt.27 This demand can, however, be supplemented by the development of private pension plans in several countries where such instruments are still incipient. Such a prospect is important because, when such a switch fails to occur or banks are otherwise unable to distribute these products, the decrease of inflows into mutual funds is likely to be accompanied by pressures on fees and the erosion of banks’ income.28

48. Competition in wholesale banking, where “market leaders” capture the bulk of the market’s earnings, may also exert pressure on the profits of large banks. The expected increase in the underwriting market, as illustrated by the sharp rise in bond issuance in 1999, may prompt more banks to enter this market and attempt to compete with the top global banks. This strategy is not without risk, however, given the resources necessary to attain the prominence required to become profitable and the unevenness of earnings that characterizes the industry. In security underwriting, for instance, the two or three lead managers of an issue typically earn more than two-thirds of the total fees, with the more numerous co-managers and the remaining syndicate members earning only a small fee each.

49. Thus, with the potential for fee income to flatten out and the variance in revenues earned from the issuance of debt and equity, banks will either need to focus on cost management to maintain profitability or, barring an ability to cut costs, banks may resort to higher risk activities.

B. Have the Economies of Scale and Scope Materialized after Consolidation?

50. Consolidation has been justified chiefly by the desire to realize economies of scale and scope. Among the most cited business reasons driving consolidation are: (1) compressed interest margins and the impetus to maintain profitability, which have led to the perception of mergers as a cost-reduction device; (2) the related deregulation and disintermediation trends that have led to the formation of financial conglomerates, able to provide a wide range of financial products and to rely less on traditional banking where profitability is declining; and (3) the need to adapt to fast-changing new technologies, especially Internet banking, which require expensive new capital investments to spread across a wider client base. According to this view, diversified revenue streams would stabilize income, with capital being used more efficiently. In addition, assuming these benefits were realized, the private franchise value of banks (i.e., the value of their brand name) would increase and banks would be more reluctant to “go for broke” when poor conditions occurred, potentially lowering the overall risk of the system.

51. It has also been argued that consolidation has been a defensive strategy (e.g., against takeovers), sometimes motivated by a desire to build up “national champions.”29 Consolidation in retail banking, in particular, can help build a well-recognized national “brand” (Box 4). While the push for domestic consolidation among retail banks also has some economic benefits—savings in clerical and back-office tasks—and is sometimes viewed as less risky than expansions abroad, such consolidation ultimately can strengthen the market power of the merger bank in the retail market and weaken the bank’s incentives to undertake the restructuring required to improve its efficiency. In addition, domestic consolidation has led to the creation of larger banks that could be considered “too big to fail” institutions.30 While the focus on building large national institutions through mergers before going abroad appears a natural step, this process may have inadvertently increased the value of access to the government’s financial safety net—raising moral hazard.

52. Cross-border consolidation of wholesale banking and processing activities, where economies of scale are more easily realized than in retail banking and where such activities are often not much in the public eye, has progressed with little resistance. Banks’ specialization in a number of wholesale business lines illustrates this point. Indeed, several major banks in the Euro area (as well as in the United States) have sold, or closed, their processing businesses and subcontracted with a handful of global providers. By the same token, the insurance industry is already dominated by fewer than a dozen companies, whose products are increasingly distributed by banks.31

Box 4.The Rise of “National Champions” and the Survival of Small Players

There are various views given for bank consolidation, the predominance of within-border consolidation, and the future of small banks. Consolidation can help raise bank’s profits by realizing cost savings and boosting market power. However, the strength of the association of these motivations with cross-border consolidation is being debated. Just the prospect of cross-border takeovers, for instance, can increase the contestability of banking markets, and may have prompted a rush of within-border consolidation and the emergence of “national champions.” More generally, in combination with consolidation, technological change may affect the viability of small banks, which may become mainly outlets for standardized products

Recent research suggests that consolidation can help spread “best banking practices,” while highlighting that it typically raises market power, where it entails substantial increases in concentration. Two stylized facts from this research are that (i) banks’ average cost curve appeared, at least until the 1990s, to be “U” shaped—thus giving weak evidence of economies of scale in commercial and retail banking before new technologies such as the Internet had spread; (ii) disparities in the mix of inputs and outputs suggested that inefficiencies averaged 20-50 percent in most countries, making it plausible that more efficient banks could profit from taking over less efficient ones. Consistently, several studies found that takeovers in the United States were followed by a rise in efficiency (Peristiani, 1993; and Hughes and others, 1999); on the other hand, foreign banks appeared to be less efficient than domestic banks (Mahajan and others, 1996). In Europe, improvements in efficiency appeared to be restricted largely to mergers of equally sized institutions and, notably, to cross-border acquisitions (Vander Vennet, 1997). Several studies—focused essentially on within border markets—also found that where concentration was high or increased in the wake of consolidation, the issue of banks’ market power emerged (Jackson, 1997; Prager and Hannan, 1999; Cymak and Hannan, 1998). These studies did not, however, address the contribution of possible foreign competition in making these markets more contestable.

The threat of cross-border consolidation may have motivated the rise of “national champions,” in some cases, with the tacit approval of national authorities. Domestic banks may have engaged in mergers and acquisitions as a way to fend off potential foreign competitors (Berger and others, 2000). Governments may have played a role in limiting cross-border mergers and breeding such champions because they preferred to have the largest banks in the country be domestically owned (Boot, 1999). Indeed, in at least one case in the euro area, national authorities mentioned the “national interest” among the reasons to block a cross-border merger. Top bank managers frequently argue that their bank needs to be big to avoid being acquired. Other reasons supporting national champions are the increased risks involved in pursuing cross-border consolidation before being “strong at home” due to different corporate cultures and insufficient knowledge of foreign markets, and—in some cases—the belief that cross-border consolidation within the euro area does not offer much scope for diversification, given the higher synchronicity of economic cycles across countries than across regions within selected countries. Some observers note that, as banks gain size, the value of access to the government’s financial safety net also increases (Saunders and Wilson, 1999). This notion can be detected in the lower ratings given by credit rating agencies that consider banks in isolation from possible government support.

“National champions” also respond to the desire of banks to achieve vertical integration in a time of change in the financial industry. Currently, there is great uncertainty about which activities will ultimately yield the largest margins; whether it is more profitable to be a producer of financial products (e.g., be an asset manager) or a distributor of such products. Hence, top banks—particularly in Europe, where the tradition of universal banking is strong—want to be present in both markets and develop a “brand” to increase their name recognition in an industry in which trust is very important.

In this environment, small banks may attempt to compete by outsourcing, although their outlook is uncertain. There is some agreement that the survival of small banks cannot be ruled out. It has been argued that they may act as niche players, and that outsourcing could help them reduce their costs and widen the range of products offered (Berger and others, 2000). Loan securitization could also help them explore potential advantages in loan monitoring and deal with a decline in customer deposits. The experience in other sectors suggests, nonetheless, that small firms—notably in distribution—quite often become mostly franchisees, an arrangement that may not be facilitated by the current ownership structure of many small banks in Europe.

53. Although it may be too early to tell, the experience with past mergers involving euro-area banks does not generally provide strong evidence of economies of scale or efficiency gains, mainly because of banks’ timid approach to labor cost cutting.

Although merged banks typically have vowed to reduce labor costs (the largest component of operating costs), consolidate their brick and mortar retail network, and concentrate on additional economies of scale in wholesale banking, most studies indicate that actual savings have usually been small.32 Such a result may not be surprising, since in many European countries laying off bank employees is not so simple—inflexible labor markets often limit the extent of labor shedding that can occur. Even where there are no explicit limitations on firings, important lobbies or strong bank-employee unions undertake the issue of job retention as a major objective. Additionally, in many countries, mandatory severance and early retirement packages paid to fired employees often make it more costly to reduce their ranks than to keep them. Mergers, despite their number, have not yet been characterized by large-scale layoffs and significant economies of scale.

54. Experience from the mergers of commercial banks in the United States and the acquisition of foreign investment banks by euro-area universal banks suggests that integration costs of mergers are higher than anticipated. In the United States labor market restrictions are substantially weaker than in the euro area though layoffs are not as sizeable as one might expect.33 Nevertheless, several of the mergers between “super regional” banks have not yielded the anticipated returns, while generating substantial up-front charges. Indeed, the stock prices of some of these banks have not to date fully recovered their pre-merger levels. Most of the benefits to U.S. mergers have been through revenue gains from cross-selling products, not from cost reductions. More generally, cross-border cultural differences or conflicting business cultures have impeded the definition of a clear post-merger business plan in the case of several purchases by large euro-area banks. These impediments have often inhibited banks from attempting to create retail networks on a de novo basis, but they may also manifest themselves within the bank’s management teams and in their approach to wholesale business.34 A number of failed cross-border acquisitions of American and British investment banks by euro-area universal banks provides evidence that these concerns are not inconsequential. Another reason why the benefits of consolidation may not be realized is that, even among private banks, the distribution of current and potential shareholder value and of the control rights of the new entity may be difficult to agree upon. In fact, an agreement about the appropriate exchange of shares at the inception of a proposed merger often causes potential partners to reconsider their decision.35

C. Could the Prevalence of Within-Border Consolidation Hurt Competition?

55. Although it is widely accepted that some degree of market power is desirable in the banking business, monopolistic practices are detrimental to the bank’s customers and unfair to its competitors.36 Banking services are of a special nature, stemming mainly from the asymmetry of information available to borrowers and lenders. It is generally agreed that where such information asymmetries are present lenders need to earn a rent to make their business of evaluating and monitoring borrowers worthwhile and avoid excessive risk taking. Taken to the extreme, however, monopolistic practices may result in excessively high prices and quantity rationing for customers.

56. A natural side effect of consolidation is to increase concentration. In the United States, antitrust regulations applied to banking mergers endeavor to define the relevant market for the provision of banks’ services and loans, in order to examine the merger’s effect on competition as well as any systemic implications.37 For retail banks, the relevant market is often deemed to be a local one, where the bank services households and small and medium-sized enterprises. In Europe, banking mergers are generally analyzed from a prudential viewpoint, and competition concerns are usually studied with a global marketplace approach.38 Although a global approach may be appropriate for wholesale banking, it is unlikely to be so for retail banking.39 A recent paper shows that, by U.S. antitrust standards, the concentration in the Swiss retail market following the UBS-SBC merger, which was evaluated largely on the global wholesale components of the two institutions, would have been unacceptable by a wide margin and would have called for the full divestiture of one retail network.40 Adverse effects of concentration have also been found in the United Kingdom where a recent report found evidence of collusive practices among the four largest banks which constitute three-fourths of the retail market.41 Though these examples are taken from countries outside the euro-area, an earlier study of the markets for mortgages and savings deposits in the Netherlands also showed signs of oligopolistic behavior, though deposit markets there are now thought to be more competitive.42

57. Within-border consolidation tends to exacerbate the risk of excessive concentration, though in some smaller EU-11 countries the viable size for a bank in today’s marketplace may mean only a few domestic banks are present (Table 9). In these countries, the banking industry may benefit from consolidation because extreme competition among too many banks precludes adequate profitability and weakens the quality of bank portfolios. To ensure that oligopolistic behavior among the few remaining banks does not occur, free entry of foreign banks needs to be assured—that is, the domestic market needs to remain “contestable.”

58. An additional implication of the creation of larger banks is the possible rationing of business to less affluent individuals and small enterprises (Box 5). Pricing strategies toward more profitable market segments, reductions in the availability of services (i.e. by closing branches post-merger), and a shift of focus toward bigger operations may reduce the supply of banking services to small customers. However, the evidence in the United States suggests that this need not be the case: although it is true that larger banks devote a smaller amount of resources to small customers, the probability that a small firm obtains a loan does not depend on the presence of small banks in the market.43 Intuitively, if a market niche is abandoned by big banks, other institutions may be willing to cover the gap, even if at a different price. Historically, savings banks and cooperatives have de facto played a potentially valuable role in some countries in helping fill this niche.

59. New technologies such as Internet banking could attenuate the risk of rationing, if given access to a sufficiently broad enough market. Most transactions done through the Internet cost between a fifth and a tenth of those executed in bank branches or even over the telephone (Box 6). This very low marginal cost should make the access to payments services and simple savings instruments affordable to even the less affluent.44 The high cost of setting up these services, on the other hand, points to the importance of scale.45 Hence, competition authorities could consider fostering the access to cross-border markets so that a critical mass is achieved without an unduly high concentration in individual domestic markets.

Box 5.Bank Consolidation and Lending to Small Enterprises

Large banks generally lend less to small firms than small banks do. In the United States, for instance, banks with under US$100 million in assets devote about 9 percent of their assets to small business lending, whereas banks with over US$10 billion invest only 2 percent of their assets in these loans (Meyer, 1998). This measure may, however, be misleading because it reflects a static situation. Indeed, the empirical evidence about the consequences of bank mergers to lending to small enterprises is mixed. Moreover, technology is reducing the specificity of loans to small enterprises, as well as their costs, making it easier for large banks to lend to small enterprises.

The rationale for believing that large banks would lend less to small firms is that small banks may be better at monitoring small borrowers, because lending to small firms involves soft information, i.e., information that is not easily quantifiable. This information is best used when loan officers have a high degree of autonomy of judgment, and large organizations may not afford that, relying instead more on quantitative criteria. The fact that the length of a bank/borrower lending relationship, which may proxy for this soft information, affects the quantity of credit offered, but not the price of credit, is indeed consistent with this intuition about small banks (Petersen and Rajan, 1994).

Empirical evidence in the United States is not conclusive about bank consolidation has being accompanied by less access to short-term credit to small enterprises.Jayarantne and Wolken (1999) find that small enterprises in areas with few small banks do not have less access to credit, i.e., bank consolidation does not cause harm to small enterprises in the long run. In the short run, a reduction in the number of small banks in a given market tends to be followed by a decline in credit lines to small firms. This reduction may not be costly for firms, however, as evidence shows that small enterprises in these markets do not switch to more expensive sources of credit such as trade credit. Berger and others (1998) noted that cross-border acquisitions (i.e., across state borders) did not result in a contraction in loans to small enterprises, and—in the case of mergers, reductions were typically offset by an increase of lending by other banks. Peek and Rosengren (1998) also highlighted the cases in which small banks were acquired by a large bank specialized in loans to small enterprises, which typically resulted in the acquired bank increasing this type of loan.

Focarelli, Panetta and Selleo (1999) show that loans to small enterprises in Italy tended to decline following the bank acquisitions, but this result deserves some qualifications. First, the finding that acquired banks had an unusually high probability of being in financial trouble—suggesting that part of the eliminated loans had negative net present value. In addition, it should be noted that the usual story about informational efficiency of small banks does not find much support in Italy, where most firms deal with a large number of banks limiting the ability of any bank to understand the credit risk of the borrowing firm—which explains the prominent role of collateral.

Turning ahead, scoring models can reduce the potential disadvantages of large banks in supplying loans to small enterprises and encourage lending to small firms (Meyer, 1998). Scoring models use statistical analysis of past credit history of the debtor and a set of financial indicators, which lead to more uniformity, lower costs, and greater potential for the securitization of small loans (they facilitate securitization by providing more of the information needed to form homogeneous pools of loans). Scoring may, of course, hurt start-up firms without a history, but they also can encourage lending to existing small firms by reducing the importance of soft information to which only one bank has access, thus permitting more banks to compete for this business. For this purpose, however, dissemination of credit information is essential. In the United States, this has been achieved through the private sector (banks can buy into the “Fair-Isaacs” credit scoring system), while in Europe central banks in Italy and France, for instance, have taken the lead.

Also, other mechanisms are increasingly substituting for traditional bank loans to small firms. Banks still have a leading role in the provision of short-term unsecured credit to firms, but a host of bank and non-bank providers now offer asset finance (e.g., leasing). Credit terms for asset finance tend to be driven more by the nature of the asset (for example, resale values) than by the characteristics of the borrower, making asset finance well suited to small enterprises looking for the financing of standard equipment and it is cheap to the provider.

On balance, financial deregulation and technology change are likely to help avoid rationing of credit to small enterprises. Credit is only one component of the panoply of services that large banks can offer to small enterprises, and a reduction in the cost of supplying it (e.g., through e-leasing and e-scoring), coupled with increasing opportunities for cross-selling, are actually likely to shift the supply of loans to small enterprises outward.

D. How are the Existing Ownership Structures Affecting the Financial Landscape?

60. Nonincorporated institutions (savings, mutual, and cooperative banks) are a distinctive feature of the euro area banking landscape. These institutions have played an important historical role, and are still key players in the provision of banking services to some segments of the market. These institutions are often credited with promoting savings and investment within their community and are still the backbone for lending to small and medium enterprises in several euro area countries. However, the large market shares of these unincorporated institutions in some countries, combined with their (past or present) privileges and their “nonprofit” status, have sometimes enabled them to pursue a market share strategy that has put pressure on commercial banks’ profits. Although some of these institutions have in recent years shifted toward a more profit-oriented strategy, their specific ownership structure may hinder their adaptability in the face of the transformation of the banking system in Europe (Box 7).

61. The large market share of public banks and other institutions that are “not for sale” may preclude consolidation within the euro area that would otherwise occur, potentially limiting economies of scope and scale in the retail sector. For example, savings banks often have no share capital, thus no shareholders, which limits merger activity. More generally, in the case of cooperatives and mutual institutions, equity “shares” are normally in the form of social participations and are not transferable, or there may be limitations on the size of participations. Thus, although these institutions have actively participated in the process of consolidation by merging among themselves or purchasing commercial banks from their store of retained earnings, their “not-for-sale” status may have prevented operations that would have otherwise taken place.46 In addition, while individually small in most countries, their prominent position overall in retail markets raises costs of setting up a branch network for outsiders, and this acts as a powerful barrier to entry. Indeed, the extensive presence of nonincorporated banks in several countries may have restricted foreign penetration by reducing the potential market left for the few banks open for acquisition.47 Moreover, in some countries, the business opportunities offered to foreign investors by privatizations were ultimately closed by the government’s distaste for accompanying restructuring plans that called for reductions in staff following the acquisitions.

Box 6.Internet Banking in Europe

Parallel to the development of the Internet, electronic banking has exploded recently in Europe. Following the lead of American and Nordic banks, where both Internet usage and Internet banking are more developed,1 virtually all major European banking groups have launched or announced in 1999 large investment programs or alliances with major telecommunications groups or Internet portals to develop this distribution channel.

Several features of Internet banking make it an attractive distribution channel.

Cost advantage: The costs of Internet banking are estimated to be as low as one-tenth of those associated with traditional distribution channels.2 These savings arise from: (i) lower overall cost of electronic transactions; (ii) greater cost economies arising from centralized information collection and transaction processing; (iii) rationalization of financial services production and standardization of banking processes; and (iv) cross-selling of nonbanking products.

Enhanced revenue opportunities: New lines of business are possible from capitalizing on existing customer bases, which can be used to match individuals with business customers for e-commerce opportunities or selling advertisement space.

Customer benefit: Customers derive several advantages, including: (i) improved price transparency as a large amount of financial services suppliers become accessible on the Internet; (ii) round-the-clock access to banking services; (iii) faster transactions where time is valuable, such as stock market operations or transfers between accounts; and (iv) easy-to-customize banking services.

Internet banking thus represents a potential strategy for large banking groups to compete in areas where the high initial cost of traditional brick and mortar branches and the dominant position of local players have traditionally acted as barriers to entry. In fact, several banks have publicly opted for this strategy for their foreign expansion (for example, Bankinter in Europe) or plan to concentrate on Internet banking to enhance their retail operations. The potential of this market has led to the appearance of pure virtual banks (such as First-e in Ireland or Telebanc in the United States) and consolidation has already started: BBVA’s Uno-e and First-e merged in 1999 to create a global Internet financial sector supermarket.

One key factor in the ability of banks to pursue this business will be the capacity to finance the large investments in information technology and marketing necessary for a successful Internet branch. In a context of high volatility of consumer sentiment, declining margins, and large probability of price wars (Uno-e has started to offer high yield savings accounts to capture market share), institutions that do not have access to capital markets to raise the necessary funds or cannot engage in share exchanges to merge with other institutions may be left behind in the race.

1 Up to 25 percent of Merita-Nordbanken customers use Internet banking, compared with 5 percent of BBVA’s.2 See Crédit Agricole (Chevreux), February 2000.

Box 7.Policy Strategies for Savings and Cooperative Banks

The strategy of euro-area governments with respect to savings and cooperative banks has followed two broad schemes:

Homogenize the banking sector by removing idiosyncrasies and privatizing savings and mutual banks, thus letting the market decide about the road to consolidation. The first strategy has been followed by Italy, which is probably the country that has undergone the deepest and farthest-ranging reform. In 1990, the Amato Law provided the legal basis for the incorporation and later consolidation of the hundreds of cooperative and savings banks. Ten years later, many of these institutions have become part of larger banking groups that are now listed on the stock exchange and participate fully in the process of consolidation (for example, Banca Intesa is the result of the merger of, among others, Cariplo, a large savings bank, and Banco Ambrosiano Veneto, a commercial bank).

Reorganize the sector from within, by encouraging mergers among savings or mutual banks but keeping the segmented nature of the market. This strategy has been followed in several other countries. In France, cooperative banks are concentrated into four large groups. In addition, the savings banks were recently transformed into the fifth cooperative group with a pyramidal structure controlled by depositors, local governments, and the state-owned CDC. The savings banks’ monopoly distribution of the Livret A was maintained. In Spain, consolidation in the savings banks has been achieved through regional mergers (for example, CajaSur is the result of the merger of the six Andalucian savings banks). In Germany consolidation of the savings banks (Sparkassen) has been ongoing for a number of years across municipalities, with some more recent consolidation in the state central banks (Landesbanken) as well. In the Netherlands, Rabobank is the result of the merger of many small cooperative banks and is now the third largest Dutch banking group.

Looking forward, it may be worth reassessing whether the historical reasons for the existence of these institutions are still present, and whether these institutions are still performing a public service deserving their special status. Recent experience in this area points to two related issues:

First, the “regional character” of these institutions is unlikely to be preserved permanently in any event, insofar as consolidation of savings and cooperative banks, even if restricted to occur among themselves, will most probably result in nation-wide institutions. Moreover, unless regional banks acting both as distributors of savings instruments and as originators of small loans are viable per se, the public ownership of such banks could require fiscal subsidies.

Second, changes in their charter need not end the funding of the social causes they may have supported. The regulation of “foundations” that owned banks in Italy actually increased the social role of these institutions—foundations will be forced by law to spend a significant share of their income in well-specified social activities—while making them more transparent and accountable for their actions. Even the social objective reflected in the provision of basic savings or payments instruments by savings banks can be achieved by allowing their distribution by the entire banking sector (the so-called banalization), or specific incentives (e.g., the targeting of the provision of free checks and basic services to some segments of the population).

62. Nonincorporated institutions may also represent a barrier for new technologies to efficiently reach small customers—the major mission for which these institutions were originally created. As mentioned, technologies such as the Internet have very low marginal cost and are best used when providers have access to a broad market—possibly across borders. The sharp reduction in costs can increase the access to basic financial products to even larger shares of the population, as long as the market is not unduly segmented. This makes it desirable to consider how the ownership and governance structures associated with the large segment of nonincorporated retail banks in Europe may affect the most efficient use of technological advances. In this regard, where cash-rich nonincorporated institutions attempt to defend their (domestic) markets by building, individually, their own platforms without the appropriate scale, the economies of the Internet could be largely wasted. There are, however, some counter examples in which some institutions have gained, mainly through alliances with similar institutions, the geographic presence and the resources to use the technology effectively. On the other hand, if online branches of large domestic or foreign banks enter the traditional market of nonincorporated banks, the possible loss of customers and the erosion of their profitability may encourage some of them to make up the lost business by taking on more risk. To the extent that nonincorporation inhibits exit, these institutions’ difficulty in adjusting is likely to ultimately result in higher costs than those incurred from the folding or absorption of uncompetitive private banks.

63. In addition, continuous consolidation of nonincorporated institutions into very large banks could lead to “too big to govern” banks—or at least banks that will be challenged by governance issues given the absence of the checks and balances imposed by publicly quoted shares or strategic shareholders. The purchase of commercial financial institutions by the non-incorporated institutions and these institutions’ desire to increase market share have led them to move away from their traditional markets (in terms of both geography and product offerings). These institutions have entered new markets and now provide services that often have little connection with the initial objectives of their charter. Further, this may lead them to take on risks that are not fully appreciated by their controlling body. In Germany, France, and Spain, for instance, a process of strict internal control and a conservative management style have so far helped prevent major failures of these types of institutions. However, such controls may not prove to be as effective when whole new activities are added to the group. For instance, assemblies of local managers may have superior insight into the appropriate strategy for a mutual bank in their retail or small business lending activities, but may have less experience in deciding the goals, focus, or methods of risk management for recently acquired investment banks. Unlike commercial banks, nonincorporated banks may not benefit from the outside signals provided by changes in stock prices, following announcements of new strategies, that convey indications as to the wisdom of certain policies (especially where cross-shareholdings do not dampen the effectiveness of these signals).48 The fear of being taken over by another institution (and the reaction to it) is muted in the case of nonincorporated banks, which are sheltered from becoming an acquisition target of private banks. The diffusely-held and non-traded stakeholdings of many mutual institutions make it particularly difficult for these owners to provide their management with appropriate signals for change.49

64. The changes afoot in the euro area suggest that unincorporated banks may need to evolve faster to weather these changes and—although there is no current evidence that these banks represent an immediate threat to the stability or soundness of the euro area banking system—their economic and social role could be usefully re-examined. Balance sheets and annual reports (where they exist) do not present information in a form that allows an in-depth analysis of unincorporated banks and thus it is difficult to tell whether the public benefits sometimes ascribed to this ownership structure outweigh their costs (e.g. preferential savings vehicles, subsidized capital, tax status, government guarantees, cumbersome governance structures). Given the speed with which banking structures in Europe are changing, countries with a large proportion of their banking systems within these structures would do well to reevaluate whether the original goals and social benefits of these banks continue to be a necessary part of the financial intermediation process and, critically, whether the slant to the level playing field in the case of publicly-owned institutions is serving a public purpose. This evaluation should pay careful attention to whether a social benefit arises from the existence of a large number of relatively small banks that could, potentially, act as a buffer in the event of a shock to the banking system as a whole. Additionally, consideration should be given to how to ensure that these institutions’ governance structures permit them to respond in economically appropriate ways to the technological and market structure changes coming in the future: alterations in their licensing or charters may need to be considered.

E. How does the Changing Financial Landscape Affect Financial Stability and Soundness of the Banking System?

65. The trends portrayed above present a euro-area financial landscape that is evolving along three main lines: integration of markets, consolidation of institutions both across product markets and, to a growing extent, across countries, and creation of very large firms that may pose systemic risks. These structural changes have potentially profound implications for the financial stability and soundness of the financial system that are analyzed below.

66. The growing integration of euro-area interbank markets—in which large banks are playing a prominent role—may increase the risk of propagation of a liquidity crisis in the region.50 The integration of the interbank market, reflected in an increase in the volume of transactions and a narrowing of spreads across countries, may increase systemic risk because this market—and especially overnight transactions—is dominated by unsecured transactions.51 These unsecured transactions exceed half of the total interbank market activity in the euro area and make up three-quarters of overnight transactions52. Opposing this development, the elimination of exchange rate risk, deeper and more liquid money markets, the gradual increase in the share of secured transactions in interbank markets in recent months (as problems such as complicated cross-border settlement procedures are addressed), and the increasing reliance of euro-area financial institutions on real time gross settlement payment systems (e.g., the TARGET system) are three positive factors enhancing the risk profile of the area as a whole.53 Parallel to market integration, the consolidation in the banking sector is likely to increase the already disproportionate share of large banks in the interbank market—especially involving cross-border transactions—thus concentrating payments in a few banks.54 The increase in size of the linked institutions may amplify the effects of the propagation of idiosyncratic shocks: although the probability of an event large enough so as to distress a larger institution is lower, if such large enough shock were to occur, it would likely be transmitted to other large institutions and potentially disrupt the payments system. Moreover, a common market event could cause more than one institution to experience difficulties at the same time. Thus, the potential failure of a few large institutions would be more difficult to resolve than if similar exposures were held among a larger number of smaller institutions. Although a case could be made that coordination problems would also make the resolution of a crisis involving many small banks difficult. All in all, the combination of market integration and banking consolidation leads to a convergence of payments risks and bank soundness concerns.

67. Consolidation, conglomeration, and extension into new business lines mean that the nature of systemic risk is changing in the euro area, making it more difficult to evaluate the possible repercussions of adverse shocks to financial stability. The effects of consolidation on aggregate risk are ambiguous. On the positive side, consolidation through cross-market and international diversification may lead to welfare enhancing risk-sharing owing to the imperfect correlation of economic cycles and asset returns across markets and countries. On the negative side, recent empirical work for banks in the United States suggests that consolidated institutions tend to shift their portfolios toward higher risk/higher expected return investments.55 Additional work extends these earlier results and finds a positive relationship between size and risk taking for the United States, Europe, and Japan.56 Higher individual risk profiles can, depending on the position of other market participants, add to the aggregate risk on the market. Consolidation has also been accompanied by an acceleration towards financial conglomeration, including through mergers and affiliations that combine banking and non-banking institutions. This trend renders the evaluation of the risk profile of banks more complex and bank supervision more difficult and costly. Banks have entered into a number of wholesale activities as well, and have stretched their retail activities to cover new products and services. The expansion into these new product areas may result in the extension of the de facto reach of the financial safety net beyond traditional deposit insurance, especially as the number of “too big to fail” institutions increases.57 More recently, the aggressive entry of some banks into e-commerce and their alliances with telecom companies and internet service providers, has compounded this risk by increasing the exposure of these banks to payments risks.58

68. The concern about the creation of more “too big to fair” institutions is especially relevant for individual euro-area countries: consolidation has created institutions whose liabilities represent a significant fraction of a country’s GDP, potentially complicating the resolution of troubled banks. Given that the type of liabilities covered by existing national deposit insurance schemes is limited to certain types of deposits and amounts, and increasingly large banks are funding themselves in other ways, the task of rescuing an insolvent institution is to a greater degree falling on the national authorities.59 The current trend toward the formation of big national champions thus implies that the failure of a major banking group may represent a significant fiscal liability for individual countries. A typical example of this issue is that the liabilities of one of the biggest banks in a small euro-area country exceeds 100 percent of GDP and the bank’s total equity represents five percent of GDP (Table 19).60

Table 19.Tier 1 Capital and Total Assets of the Largest Banking Group(In percent of GDP)
Tier 1 CapitalTotal AssetsTier 1 CapitalTotal AssetsTier 1 CapitalTotal Assets
United Kingdom1.0820.711.6028.002.3339.75
Sources: The Banker, various issues; and IMF, International Financial Statistics.
Sources: The Banker, various issues; and IMF, International Financial Statistics.

69. On balance, these trends raise serious issues regarding three key aspects of financial oversight—supervisory arrangements, crisis management, and resolution techniques—and calls for vigilance regarding the impact of the predominantly domestic consolidation. It is sometimes suggested that the integration of interbank markets, the quickening of the pace of bank consolidation, and the creation of conglomerates spanning several types of financial services are trends that are evolving independently, and as such could be taken as evidence that the prevalence of domestic mergers is not hindering the appropriate development of more efficient and safe financial markets in Europe. However, there are at least some reasons to think otherwise. It is clear that the benefits of integrated markets can proceed in absence of cross-border mergers. However, the development of very large domestic banks in an environment of increased linkages and in view of the present supervisory structures, crisis management, and resolution techniques, can not only hinder competition in domestic markets, but also pose challenges for effective financial oversight. For instance, excessively big domestic institutions may increase the incentives for forbearance, complicate the response at the euro level to any liquidity crises, and weaken the credibility of deposit insurance schemes. Indeed, they can slow the movement toward better coordination of supervisors across the region since the incentives remain domestic: the local political repercussions of not being able to resolve a solvency or liquidity problem at home before it becomes systemic may preclude prompt notification of other supervisors.

F. Are Current Supervisory Structures Keeping Pace with the Changing Landscape?

70. The previous section has examined the possible effects on systemic risk and financial stability of the ongoing trends underlying the transformation of the euro area banking system. This raises an obvious question: is the current supervisory structure adequate to accommodate an evolving banking system that may become more concentrated, more integrated, and more capital market oriented over time?

71. The current banking supervisory structure in the euro area is organized along national lines and, as a result of the introduction of the euro, the geographic domain of monetary policy no longer coincides with that of prudential supervision. Each country has a bank supervisory organization that in almost all cases has either direct links with the central bank (Box 8) or strong operational connections. Insurance supervision is usually undertaken by a separate institution, and securities supervision is often connected to the banking supervisor, but may exist separately. These arrangements reflect a variety of historical developments and philosophies regarding the appropriate role of regulation and supervision. Bank supervision, for instance, typically gravitated initially toward central banks—sometimes through the delegation of responsibilities from infra-national authorities—largely because of the role of central banks as ultimate providers of liquidity and guarantors of financial stability. The separation of the central bank in the euro area from the supervisory institutions, in conjunction with the shift from a national to a supranational monetary policy, implies that supervisors need to be more vigilant, to the extent they can no longer rely upon the central bank to lower interest rates to forestall the spread of a domestic systemic crisis due to a weak domestic banking system. As well, nor can supervisors rely on their central bank to raise rates to dampen credit overheating. Some EU authorities have argued that there is no evidence that central banks would consider these avenues. Others have noted that the separation of banking supervision and monetary policy, nevertheless, helps ensure that an inflationary bias to monetary policy is removed.

Box 8.Banking Supervision in the European Union

The supervision of the banking sector in the European Union is based on the EU’s Second Banking Directive. Building on the principles of mutual recognition and home country control, domestic banks, subsidiaries of foreign banks, and branches of non-EU banks are supervised by the domestic authorities, whereas branches of EU banks are supervised by the home country authorities.

CountrySupervisory Institution
AustriaMinistry of Finance
BelgiumBanking and Finance Commission
FinlandFinancial Supervision Commission
FranceBanking Commission
GermanyFederal Banking Supervisory Office
IrelandCentral Bank of Ireland
ItalyBank of Italy
LuxembourgFinancial Sector Surveillance Commission
NetherlandsNetherlands Bank
PortugalBank of Portugal
SpainBank of Spain

The ongoing consolidation drive in the European banking sector raises the important issue of the supervision of subsidiaries of foreign banks, and the need to exchange information with home country supervisors to achieve a consolidated view of the group to which the subsidiary belongs.

In this respect, the current practice in the EU as regards practical cross-border supervisory arrangements generally comprises memoranda of understanding signed with foreign supervisory authorities, and exchanges of information at annual meetings. In addition, for countries within the EU, the Groupe de Contact—a group of European supervisors that meets three times a year to discuss supervisory practices and individual banks’ cases—provides a forum for institutionalized cooperation and exchange of information.

More related to policy advice and macroprudential surveillance, two additional committees exist at the European level:

  • The Banking Advisory Committee, established in the First Banking Directive to advise the Commission on broad supervisory policy issues. It is composed of a representatives of the central bank, the ministry of finance and the supervisory authority of each EU country and three representatives of the EC.
  • The Banking Supervisory Committee at the European Central Bank, which advises its Council on issues falling within the competence of national central banks and affecting the stability of financial institutions and markets.

72. Above the level of national authorities is an EU-level structure that has provided a framework through several directives, based on the principles of mutual recognition and home country control. Deposit insurance also follows the home country rule, though a minimum amount of coverage has been harmonized across EU countries (Table 20). There has been agreement within the Euro system regarding a policy for emergency liquidity assistance in which its provision, if and when it is deemed appropriate, is primarily a national responsibility.61 It is believed by members of the Euro system that adequate institutional mechanisms are currently in place to ensure that the appropriate information is exchanged among national authorities and the ECB so that the impact of liquidity provision can be managed without undue disruption to the monetary policy stance.

Table 20.Status and Cover Limits of the Deposit Guarantee Schemes in the European Economic Area
CountryStatusCompensation Amount 1/
EuroNational currency
Austriaprivate20,000 5/
Belgiummixed (public/private)20,000 2/
Denmarkprivate40,000DKr 300,000
Germanyprivate/equivalent20,000 (90%) 3/
Icelandprivate20,000ISK 1,700,000
Irelandpublic20,000 (90%) 2/
Liechtensteinprivate19,000 4/Sw F 30,000
Luxembourgprivate20,000 2/
Norwaypublic/private250,000NKr 2,000,000
SwedenPublic25,000SKr 250,000
SpainMixed20,000 2/
United KingdomPublic22,000 (90%)£20,000 (90%) 6/
Sources: Bundesbank Monthly Report, July 2000, data from European Commission, COM (1999) 722, of December 22, 1999.

For specific regulations on the amount of compensation and on the scope of protected liabilities and/or protected creditors, please refer to the national compensation scheme concerned.

Compensation from January 1, 2000; €15,000 up to December 31, 1999.

90 percent of the deposits, maximum €20,000; institutional protection by the Federal Association of German People’s Banks and Raiffeisen Banks, and German Savings Bank and Giro Association; supplementary cover by voluntary schemes operated, in particular, by the Federal Association of German Banks, the Federal Association of Public Banks, the private building and loan associations.

No Commission data in euro; ECB euro reference exchange rate at end-April 2000.

90 percent of the deposits of legal persons are covered.

90 percent of the deposits up to a maximum amount of £20,000.

Sources: Bundesbank Monthly Report, July 2000, data from European Commission, COM (1999) 722, of December 22, 1999.

For specific regulations on the amount of compensation and on the scope of protected liabilities and/or protected creditors, please refer to the national compensation scheme concerned.

Compensation from January 1, 2000; €15,000 up to December 31, 1999.

90 percent of the deposits, maximum €20,000; institutional protection by the Federal Association of German People’s Banks and Raiffeisen Banks, and German Savings Bank and Giro Association; supplementary cover by voluntary schemes operated, in particular, by the Federal Association of German Banks, the Federal Association of Public Banks, the private building and loan associations.

No Commission data in euro; ECB euro reference exchange rate at end-April 2000.

90 percent of the deposits of legal persons are covered.

90 percent of the deposits up to a maximum amount of £20,000.

73. In addition to these national structures, the Basel Committee on Banking Supervision, consisting of supervisors from the major developed countries, promulgates and encourages sound banking supervisory practices. The EU has adopted most of the Basel Committee’s practices through various directives which must be implemented through national law and regulation within the individual EU countries. Similar organizations in the securities and insurance sectors, the International Organization for Securities Commissions (IOSCO), and the International Association of Insurance Supervisors (IAIS) also provide a backdrop for euro-area financial sector supervision.

74. The trend toward larger, more diversified, and more internationally oriented institutions raises important issues about the current arrangements in which supervision is segmented across product markets. Several European countries, including Ireland and the United Kingdom, are in the process of or have already formally revamped their supervisory structure along these lines. France is also considering further integration. The experience so far with different arrangements is not conclusive, as many of these changes have only recently been implemented. Overall, the crucial aspect for the efficient supervision of financial conglomerates is the clear definition of and agreement on the “lead supervisor” (Box 9). Designing an effective “lead supervisor” model is compatible with both unified and decentralized supervision, provided that adequate coordination mechanisms are in place and there is alignment of incentives and objectives across agencies.

Box 9.Single Financial Supervisors or Specialized Supervisors?

A decentralized supervisory architecture at the euro area level raises the debate of the optimal institutional structure at the national level. While supervision in euro area countries remains largely separated by sector (see table), some EU countries, such as the United Kingdom, Finland, Sweden, and Denmark, have moved to an integrated supervisory authority, in which the different agents of the financial sector (banks, insurance companies, and financial markets) are supervised by a single agency. In some of these countries (and also in Korea and Japan), the reform was the outcome of perceived weaknesses in supervision after a number of financial failures, but were also justified by the ongoing blurring of traditional boundaries between financial activities.

Source: Updated from Lannoo (1999).Notes: BS=banking and securities; I=specialized insurance supervisor; G=government department; CB=central bank; B=specialized banking supervisor; S = specialized securities supervisor.
Source: Updated from Lannoo (1999).Notes: BS=banking and securities; I=specialized insurance supervisor; G=government department; CB=central bank; B=specialized banking supervisor; S = specialized securities supervisor.

Several arguments can be proposed for and against both models of supervision (see, for example, Goodhart and others (1997) and Lannoo (1999)). The case for a unified supervisory authority relies on several facts: (i) it facilitates the comprehensive assessment of risks and may generate economies of scope, by pooling the expertise of different functional supervisors and guaranteeing their cooperation; (ii) it would create a one-stop agency for authorizations and settlement of complaints, and may reduce transaction costs and supervisory fees; (iii) it may enhance transparency and accountability, to the extent that the allocation of responsibilities for supervision of financial conglomerates is clear and undivided; and (iv) by reducing the number of authorities and homogenizing their structure, it may facilitate cooperation among European supervisors.

The case against a unified supervisory authority builds on: (i) the risk profile and nature of business is substantially different across sectors, and excessive homogenization across heterogeneous activities may decrease the overall quality of supervision, casting doubts on the likelihood of achieving economies of scope from joint supervision; (ii) a very powerful supervisor could increase moral hazard if the public perceives the financial system to be under control, thereby reducing the incentive for financial institutions to prudently manage their own business; (iii) supervision is becoming increasingly specialized, especially as more emphasis is given to market discipline in risk control; and (iv) a single authority would eliminate the potential benefits from regulatory competition.

Overall, the crucial aspect for an efficient supervision of financial conglomerates is the clear definition of and agreement on the “lead supervisor.” The experience so far with unified supervision is not conclusive, as all of these agencies have been recently created. Designing a “lead supervisor” is compatible with either model of supervision, provided that adequate coordination mechanisms are in place.

75. Additional burdens on the decentralized model for banking supervision arise when supervising the increasing number of internationally active financial conglomerates. Though the mechanisms differ, several countries are trying to improve their oversight in this area. France, the Netherlands, Spain, and the United Kingdom have recently set up special schemes for the in-depth supervision of their largest global banking groups, and the forthcoming EU directive on financial conglomerates may reinforce this process. In the United States, the Federal Reserve Board issued a special letter for the supervision of large, complex banking organizations (LCBOs).62

76. At the multilateral level, the Joint Forum on Financial Conglomerates (Joint Forum) was spawned in early 1996 to help facilitate improved supervision of these internationally active entities. Under the aegis of the Basel Committee, the IOSCO and the IAIS, the Joint Forum has reviewed various means for facilitating information exchanges among supervisors.63 It has examined ways to enhance supervisory coordination, including the benefits and drawbacks of establishing criteria to identify and define the responsibilities of a coordinator, and is working on developing principles toward the more effective supervision of regulated firms within financial conglomerates.64 Progress has been slow and a working criterion for identifying a coordinator has not yet been made operational. These advances should aid the work of supervisors world-wide, and in particular help to develop a better defined set of information to exchange.

77. At the EU level, the recently released “Brouwer Report”65considers that existing institutional arrangements are adequate, and points to the ongoing integration in some euro area countries of supervisory structures across markets as an encouraging development. The report notes the development of an integrated euro money market and the increasing depth and liquidity of the euro securities markets—both of which contribute to financial stability. It further notes that the harmonization of financial regulation within the EU is generally consistent with the proposals and guidelines of international institutions and, in some cases, goes beyond. Importantly, while the ideal “Single Market” in the EU has not yet been achieved, the report notes that the EU directives provide many of the elements of the legal and regulatory underpinnings that are necessary for the ongoing integration of financial market structures. Though no institutional changes are thought necessary, the Brouwer Report suggests, however, some room for improvement, especially as regards strengthening cross-country supervisory coordination. In particular, it advances that (i) enhanced exchange of confidential information on a regular basis ought to be pursued; (ii) reinforced cooperation is required for increasing convergence in supervisory practices, and (iii) a strong involvement of central banks in supervisory cooperation is called for, especially to ensure smooth coordination in times of stress.

78. Despite the efforts already underway, some important impediments remain to be addressed in order for the recommendations of the Brouwer Report to become fully and swiftly effective. Critically, the information to be exchanged, itself, needs to be easily and consistently interpreted. Given the current lack of uniformity across jurisdictions on an item as important as nonperforming loans, increased attention to the comparability of information will undoubtedly become an important issue for further discussion. As part of this effort, convergence of supervisory practices is being discussed. Moreover, since many supervisors hold the view that knowledge of individual institutions is best gained in a decentralized fashion and that judging the extent of individual banking problems, such as liquidity or insolvency, can best be made by maintaining a decentralized supervisory organization, additional incentives for cooperation and communication with the European Central Bank may need to be established. Decentralized supervision is important for slow-to-develop problems with individual institutions, but this approach may be less suited to the increasing capital market orientation of large, globally connected, financial institutions where problems can arise and propagate quickly and fast responses may hold the key to a successful intervention. The quick-to-develop problems are more likely, though not exclusively, to be liquidity problems requiring a more centralized response. Decentralized supervision will continue to be vital for the early detection of solvency problems.66

79. The European Union has made considerable progress in addressing the challenges posed by these rapid changes in the financial sector. However, some hurdles remain. More specifically, the combination of a unified monetary union—with its implications for area-wide monetary policy—and the complexity surrounding responsibilities at the national and European levels raises at least four specific issues related to the prevention and management of liquidity and insolvency episodes:

  • The jurisdiction of national supervisors is likely to become more blurred as cross-border mergers become more common. The problem of ensuring a rapid flow of information in response to liquidity shocks affecting the euro area is likely to be compounded by the uncertainties regarding the responsibilities of national supervisors in the case of cross-border groups. The Brouwer Report notes that clarification and extension of the concept of the coordinating supervisor(s) for the large financial groups domiciled in Europe could be further improved. Furthermore, in the few large cross-border mergers that have already taken place, the home country rule has been usually complemented with ad hoc Memoranda of Understanding (MOUs)—which set up the operational aspects of supervision of individual cross-border groups. Although the establishment of MOUs is evidence that the parties have made ex-ante arrangements for various contingencies, their existence also reveals the lack of specificity of existing directives and legal structures to accommodate the increasing internationalization of the banking sector and the need to transfer potentially confidential information across borders. In addition, such arrangements might be more difficult to rely upon when more than two parties are involved in their negotiation.
  • As regards the management of liquidity crisis, the rapid increase in cross-border interbank credit has highlighted the implications of regional integration for the spreading of liquidity risk. Interbank credit markets have deepened considerably since early 1999, with a notable increase in reliance on unsecured cross-border loans fueled inter alia by the elimination of currency risk within the area and the very strong growth in credit in some countries. In such an integrated market, liquidity problems can spread quickly, putting a special burden on the ability of the ECB to respond quickly to local problems and—for this purpose—to identify whether a crisis reflects mainly a liquidity shortage or deeper solvency issues. One could argue that should a major banking crisis occur in the euro area, both national and euro area authorities would be immediately held responsible, regardless of whether they were able to detect or prevent its occurrence or whether they had the appropriate tools to act promptly and efficiently. Thus, it is important that ownership of the problem be allocated to those who will be held responsible. This heightens the need for close coordination between banking supervision and market surveillance, and underscores the importance of securing fast and reliable transmission of information between individual countries and the ECB.
  • The provision of emergency liquidity assistance is also subject to incentive problems that may blur the already gray line between liquidity and solvency incidents. Since National Central Banks (NCBs) can still provide liquidity to troubled institutions, it is plausible that excessive within-border consolidation could result in “too big to fail” banks. The specific language in the Statute of the European System of Central banks suggests that the NCBs have considerable leeway with respect to the type of assets they can purchase (as collateral) from a bank facing a liquidity squeeze. Unless the Governing Council of the ECB decides that such an intervention or the extension of guarantees to banks “interferes with the objectives and tasks of the ESCB,” NCBs could thus assume banks’ credit, market and liquidity risk.67 Hence, the potential for moral hazard is still present, as banks may continue to count on national safety nets, and a NCB may ultimately step in to support an otherwise insolvent bank that at the domestic level is “too big to fail,” (or at least “too big to liquidate” quickly) while the institution is of no systemic importance at the European level.
  • Finally, an eventual rescue of a very large (and possibly cross-country) insolvent institution poses the problem of the financing and ownership of the resolution operation. The current system of domestically defined deposit guarantee schemes may not have the resources to address the failure of any of the top euro area banking groups, and thus fiscal resources may have to be mobilized. In the absence of a well defined procedure, a rescue involving several countries may be difficult to coordinate and present to the public. Furthermore, even though the home country principle seems to assign the final ownership of the resolution operation to the home country supervisor, this rule may be highly problematic in cases of countries of very different sizes.68 Resolution difficulties due to institutional impediments, however, should not be used as reasons to dissuade cross-border mergers from taking place, but should be removed to make the exit of merged banks that end up being non-viable institutions as smooth as possible.

80. Thus, the heightened potential for contagion resulting from increased linkages among euro-area financial institutions and the various impediments to an efficient coordination process raises the question of a centralized supervisory unit. Recent theoretical research also points in this direction: Dell’Ariccia and Marquez (2000) show that a centralized regulator that takes into account international externalities would enforce higher standards than a decentralized system, and that the benefits of a centralized system become higher as the financial markets become more integrated. However, it is recognized that the full integration of existing supervisory organizations within the euro area would not only be particularly difficult but would also not be desirable at this stage. One of the major difficulties arises because supervisory organizations are under the jurisdiction of different governmental units, and in most cases not consolidated across banking, securities, and insurance within the domestic sphere (Box 8). Even before units could be integrated across borders, enabling legislation may be required domestically. It is also the case that in some countries different kinds of banks are supervised under different guidelines or rules due to their different roles.69 This suggests that an overarching supervisory unit would be overly cumbersome and there would be a loss of country-specific skills.

81. However, serious consideration should be given to a centralized surveillance unit with effective links to supervisors that would focus primarily on the market activities of the largest euro-area banks because of their increasingly global, capital markets orientation and their predominant role in the integrated euro area-wide payments system. Given the potential for increased systemic risk, surveillance of the positions and market flows of the largest banks—a few dozen institutions considered of systemic importance—as well as tailored supervision of their risk management systems are likely to be most efficiently performed from the center of euro-area markets. This would complement the more detailed and narrower scope of the national supervisors.70 It would further provide an opportunity for national supervisors to share their views on the vulnerability of their banking systems based on aggregate information. Thanks to a bird’s eye view on cross-border interconnectedness and exposures of euro-wide institutions, the centralized surveillance unit could effectively use this information to evaluate euro-area risks resulting from cross-border spillovers in case of crisis. It is important to maintain and enhance national supervisory units as these units are most familiar with the domestic legal frameworks, including bankruptcy laws, as well as accounting and cultural differences that influence the performance of the domestic banks. In any event, increasing complexities associated with financial institutions will require more micro and macro attention to financial sectors: constant exchange of information will likely become a necessity.

82. Such enhanced market surveillance resulting from systemic risk and market stability concerns can be performed by an independent institution. There are two characteristics of the ECB, however, that suggest that it could be well-placed to take on the role of overseer. First, the ECB already has the mandate to ensure the smooth functioning of the TARGET payments system.71 The central payments system is often an integral transmission mechanism for financial shocks and provides timely information about various institutions’ transfers to one another. Second, the ECB, de facto, would play a role in ensuring market stability though the provision of liquidity in the event of a system-wide liquidity crunch that was unable to be accommodated by the NCBs. However, such a move would not solve the existing problem regarding the fiscal burden that could be associated with bank insolvencies.72 Overall, the chosen institutional arrangements should be such that, building on the expertise of national supervisors, they provide adequate surveillance over the market activities a set of “systemically important” institutions that have the potential to disrupt markets and derail monetary policy objectives. In addition, arrangements should be carefully designed to ensure that the role of the lead supervisor is not undermined and that the legal competence and accountability of each institution are clearly defined.73

83. In the more capital-markets oriented European environment, an in-depth understanding of the links between financial institutions’ market and credit exposures is not possible without access to supervisory information. Financial market surveillance should, ideally, be able to look behind the aggregate numbers and be able to assess the riskiness of the behaviors underlying them. This is important given the increasing connections between banks’ exposures (both on- and off-balance sheet) and market movements. As well, an understanding of institutions’ risk management systems and likely responses to stress is vital for assessing potential outcomes of shocks for financial stability.

84. The emphasis in this paper has been placed on increasing the capacity to perform euro-area market surveillance complemented by better euro-wide supervisory links; arguably, though, and given the connections of major euro area financial institutions outside the euro area, the argument could be extended to advocate a global surveillance framework. While this could be considered a logical extension, the fact is that in practice, issues of competencies and implementation place this option beyond consideration at this time. Nevertheless, the ongoing revision of the international financial architecture, especially through the work taking place in international fora such as the Financial Stability Forum and the Joint Forum, clearly define a trend toward enhanced global market surveillance combined with strengthened supervisory linkages. This may be the ultimate end-point of the ongoing globalization in financial markets and of the management of contagion risks that these integrated markets now present.



The data used to compute banks’ performance indicators (Table 12) comes from the FitchIBCA database. This database compiles data from individual banks’ balance sheets and income statements from a large set of countries, according to a harmonized definition of the variables across countries. This allows us to build consistent indicator variables across different national banking systems to make their comparison possible. Since only a subset of banks within each country is represented in the database, the indicators presented in the paper may differ from those reported in central banks bulletins. However, since large banks are all included in the database, differences mainly originate from the characteristics of small banks. This bias is potentially more serious for the smaller countries. For example, in Finland, FitchIBCA data shows that the return on equity in 1998 of the banks in its universe (12 banks) as 14 percent, whereas national sources report the return on equity as 26 percent for the banking system as a whole (33 banks). The data covering Luxembourg is also problematic in this regard because of the large number of foreign-owned banks and the fact that the IBCA reports we use are presented on a consolidated basis. Corrections (and accompanying notations) are made in text tables for which the biases appear large. The table below lists the number of banks per country provided in FitchIBCA, as compared to their actual number in that country.

Table 1.Number of Banks and FitchIBCA Sample 1/
Sources: FitchIBCA; central banks; and bankers’ associations.

Banks are taken to include only commercial, savings and cooperatives banks.

Sources: FitchIBCA; central banks; and bankers’ associations.

Banks are taken to include only commercial, savings and cooperatives banks.

Table 2.Number of Institutions
Commercial banks245338331322323
Of which: Foreign6360697582
Savings banks602781637611607
Cooperative banks3,6643,3842,5952,4222,253
Total universal banks4,5114,5033,5633,3553,183
Post office19,70616,10014,702
Investment companies3454656870
Total credit institutions4,7394,7113,7853,5783,404
Commercial banks422406382
Of which: Foreign90
Savings banks353434
Cooperative banks132127124
Total universal banks589567540
Post office11111
Investment companies473503
Total credit institutions2,1052,0271,4691,2991,237
Commercial banks225283302
Of which: Foreign525559
Savings banks
Cooperative banks715652619
Total universal banks940935921
Post office11111
Investment companies
Total credit institutions1,1921,156970935
Commercial banks139154170159152
Of which: Foreign5351
Savings banks7966515151
Cooperative banks146107979797
Total universal banks364327318307300
Post office
Investment companies
Total credit institutions695696506416404
Commercial banks
Of which: Foreign812
Savings banks
Cooperative banks
Total universal banks970
Post office11111
Investment companies
Total credit institutions124112401041994970
Commercial banks949510310190
Of which: Foreign2934404039
Savings banks3128252118
Cooperative banks2424161211
Total universal banks 1/165157145134120
Post office
Investment companies
Total credit institutions165157145134120
Commercial banks1015121415
Of which: Foreign00466
Savings banks254150394040
Cooperative banks370338301294289
Total universal banks634503348348338
Post office
Investment companies4046
Total credit institutions654503352348344
Commercial banks444850
Of which: Foreign
Savings banks
Cooperative banks437
Total universal banks5157
Post office11111
Investment companies
Total credit institutions5848485157
Commercial banks209
Of which: Foreign106155193188182
Savings banks0
Cooperative banks22
Total universal banks211
Post office11111
Investment companies788083
Total credit institutions118177220215209
Commercial banks8397989894
Of which: Foreign252226
Savings banks6654262624
Cooperative banks1877878547481444
Total universal banks20261029671605562
Post office11111
Investment companies446
Total credit institutions20551058721650606
Commercial banks476062
Of which: Foreign1719
Savings banks899
Cooperative banks183170160
Total universal banks238239231
Post office111
Investment companies163165167
Total credit institutions319327297
Commercial banks483480464
Of which: Foreign262265259
Savings banks 2/807171
Cooperative banks
Total universal banks563551464
Post office11111
Investment companies
Total credit institutions564551
Commercial banks14,41712,3479,9429,1438,774
Of which: Foreign
Savings banks 3/3,6262,8152,0301,7801,687
Cooperative banks 4/12,86011,68711,238
Total banks18,04328,02223,65922,16110,461
Post office
Investment companies
Total credit institutions
Commercial banks171165167
Of which: Foreign949389
Savings banks
Cooperative banks4,6624,0063,335
Total universal banks4,8334,1713,502
Post office11111
Investment companies
Total credit institutions
Sources: National central banks; national banking associations; European Central Bank; Bank of International Settlements; OECD; European Central Bank; and Federal Deposit Insurance Company.Note: Includes other specialized financial institutions (e.g., mortgage companies, municipal credit banks, etc.)

In 1985 and 1990, it also includes 16 and 10 Law 10/06/64 banks respectively.

Building societies.

Thrift institutions.

Credit unions.

Sources: National central banks; national banking associations; European Central Bank; Bank of International Settlements; OECD; European Central Bank; and Federal Deposit Insurance Company.Note: Includes other specialized financial institutions (e.g., mortgage companies, municipal credit banks, etc.)

In 1985 and 1990, it also includes 16 and 10 Law 10/06/64 banks respectively.

Building societies.

Thrift institutions.

Credit unions.


    BaliñoT. and A.Ubide2000The New World of BankingFinance and Development (June) pp. 41-44.

    Bank for International Settlements199969th Annual Report (Basle, Switzerland).

    Bank of Italy1999Annual Report (Rome, Italy).

    BergerA.R.Demsetz and P.Strahan1998The Consolidation of the Financial Services Industry: Causes, Consequences, and Implications for the FutureJournal of Banking and Finance (23 February) pp. 135-194.

    • Search Google Scholar
    • Export Citation

    BergerA.R.DeYoungH.Genay and G.Udell2000Globalization of Financial Institutions: Evidence from Cross-Border Banking Performancemanuscript, forthcoming in Brookings-Wharton Papers on Financial ServicesVol 3.

    • Search Google Scholar
    • Export Citation

    BootA.1999European Lessons on Consolidation in BankingJournal of Banking and Finance (23 February) pp. 609-613.

    CollenderR. and S.Shaffer2000Local Bank Office Ownership, Deposit Control, Market Structure, and Economic GrowthWorking Paper TB-1886 Economic Reserve Service/United Stated Department of Agriculture (Washington, D.C.)

    • Search Google Scholar
    • Export Citation

    Crédit Agricole (Chevreux)February2000.

    CrockettA.(1997) “Why is Financial Stability a Public Policy GoalFederal Reserve bank of Kansas City Economic ReviewFourth Quarter p. 5-22.

    • Search Google Scholar
    • Export Citation

    CruickshankD.2000Review of Banking Services in the U.K.United Kingdom Government,

    CyrnakA. and T.Hannan1998Bank Lending to Small Businesses and the Use of CRA Loan Data to Measure Market StructureWorking Paper Federal Reserve Board (Washington D.C.)

    • Search Google Scholar
    • Export Citation

    De NicolóG.2000Size, Charter Value and Risk in Banking: An International PerspectivemimeoBoard of Governors of the Federal Reserve SystemUnited States.

    • Search Google Scholar
    • Export Citation

    Dell’ AricciaG. and RobertMarquez(2000) “Competition among RegulatorsmanuscriptInternational Monetary FundWashington, D.C.

    • Search Google Scholar
    • Export Citation

    Deutsche Bundesbank2000Deposit Protection and Investor Compensation in GermanyMonthly Report (July).

    European Central Bank (ECB)1999aThe Effects of Technology on the EU Banking Systems” (July).

    European Central Bank (ECB)1999bTarget and Payments in EuroMonthly Bulletin (November).

    European Central Bank (ECB)2000aEMU and Banking SupervisionMonthly Bulletin (April).

    European Central Bank (ECB)2000bThe Euro Area One Year After the Introduction of the Euro: Key Characteristics and Changes in the Financial StructureMonthly Bulletin (January).

    • Search Google Scholar
    • Export Citation

    European Central Bank (ECB)2000cThe Impact of the Euro on Money and Bond Markets” by J.SantillánM.Bayle and C.ThygesenOccasional Paper No. 1 (Frankfurt).

    • Search Google Scholar
    • Export Citation

    European Commission2000Report on Financial Stability(the Brouwer Report), Economic and Financial Committee, EFC/ECFIN/240/00 (Brussels).

    • Search Google Scholar
    • Export Citation

    European Commission1997The Single Market Review: The Impact on Services: Credit Institutions and Banking” (Luxembourg).

    European CommissionFinancial Services Legal NoticeOctober201999.

    Federal Reserve System Board of Governors1999Risk-Focused Supervision of Large Complex Banking OrganizationLetter to the Officer in Charge of Supervision and appropriate supervisory and examination staff at each federal reserve bank etc… (June231999) SUP–SR 99-15.

    • Search Google Scholar
    • Export Citation

    Financial TimesEuropean Banks Rule Out German Takeover Bids” (April122000).

    FocarelliD.F.Panetta and C.Salleo1999Why do Banks Merge?Banca d’Italia, Temi di discussione 361.

    GoodhartC.P.HartmannD.LlewellynL.Rojas Suarez and S.Weisbrod1997Financial Regulation: Why How and Where Now? (Routledge).

    • Search Google Scholar
    • Export Citation

    HawkesbyC.(1999) “Central Banks and Supervisors: The Question of Institutional Structure and Responsibilities” in Financial Stability and Central Banks – Selected Issues for Financial Stability and Market DisciplineCentre for Central Banking Studies, Bank of England.

    • Search Google Scholar
    • Export Citation

    HughesJ.W.LandL.Mester and C.Moon1999The Dollars and Sense of Bank ConsolidationJournal of Banking and Finance (23 February) pp. 291-324.

    • Search Google Scholar
    • Export Citation

    HoustonJ.C.James and M.Ryngaert1999Why are Value Enhancing Mergers in Banking So Hard to Find?Gainesville Department of Finance, Insurance, and Real Estate, University of Florida.

    • Search Google Scholar
    • Export Citation

    International Monetary Fund1999International Capital Markets: Developments, Prospects, and Key Policy Issues” (Washington D.C.)

    • Search Google Scholar
    • Export Citation

    JacksonW.1997Market Structure and the Speed of Price Adjustments: Evidence of Non-MonotonicityReview of Industrial OrganizationNo. 12(1) pp. 37-57.

    • Search Google Scholar
    • Export Citation

    JayrantneJ. and J.Wolken1999How Important are Small Banks to Small Business Lending? New Evidence from a Survey to Small BusinessJournal of Banking and Finance (23 February) pp. 427-458.

    • Search Google Scholar
    • Export Citation

    KwastM.M.Starr-McCluer and M.Wolken1997Market Definition and the Analysis of Antitrust in BankingAntitrust BulletinNo. 42 pp. 973-995.

    • Search Google Scholar
    • Export Citation

    LannooK.1999Financial Supervision in EMUYearbook of International Financial and Economic Law pp. 146-171.

    MahajanA.N.Rangan and A.Zardkooki1996Cost Structures in Multinational and Domestic BankingJournal of Banking and Finance (20 February) pp. 238-306.

    • Search Google Scholar
    • Export Citation

    MeyerL.1998The Present and Future Roles of Banks in Small Business FinanceJournal of Banking and FinanceNo. 22 pp. 1109-1116.

    • Search Google Scholar
    • Export Citation

    NevenD. and T.von Ungern-Sternberg1998The Competitive Impact of the UBS-SBC MergerHEC University of Lausane working paper.

    • Search Google Scholar
    • Export Citation

    OECD1998Enhancing the Role of Competition in the Regulation of Banks (Organization for Economic Cooperation and DevelopmentParis).

    • Search Google Scholar
    • Export Citation

    PeekJ. and E.Rosengreen1998Bank Consolidation and Small Business Lending: It is Not Just Bank Size that MattersJournal of Banking and FinanceNo. 22 pp. 799-819.

    • Search Google Scholar
    • Export Citation

    PeristianiS.1993The Effects of Mergers on Bank PerformanceFederal Reserve Bank of New York Studies on Excess Capacity in the Banking Sector (March).

    • Search Google Scholar
    • Export Citation

    PetersenM. and R.Rajan1994The Benefits of Firm-Creditor Relationships: Evidence from Small Business DataJournal of FinanceNo. 49 pp. 3-37.

    • Search Google Scholar
    • Export Citation

    PragerR. and T.Hannan1999Do Substantial Horizontal Mergers Generate Significant Price Effects? Evidence from the Banking IndustryJournal of Industrial Economics.

    • Search Google Scholar
    • Export Citation

    PratiA. and G.Schinasi1999Financial Stability in European Economic and Monetary Union Princeton Studies in International FinanceNo. 86 (Princeton UniversityPrinceton, N.J.)

    • Search Google Scholar
    • Export Citation

    SaaparI. and SoussaF.(1999) “Financial Consolidation and Conglomeration: Implications for the Financial Safety Net” in Financial Stability and Central Banks – Selected Issues for Financial Stability and Market DisciplineCentre for Central Banking Studies, Bank of England.

    • Search Google Scholar
    • Export Citation

    SaundersA. and B.Wilson1999The Impact of Consolidation and Safety-Net Support on Canadian, U.S. and U.K. BanksJournal of Banking and FinanceNo. 23 pp. 537-571.

    • Search Google Scholar
    • Export Citation

    ShafferS.1993Can Megamergers Improve Bank Efficiency?Journal of Banking and FinanceNo. 17 pp. 423-436.

    SoussaF.(1999) “Too Big to Fail: Moral Hazard and Unfair Competition” in Financial Stability and Central Banks – Selected Issues for Financial Stability and Market DisciplineCentre for Central Banking Studies, Bank of England.

    • Search Google Scholar
    • Export Citation

    SwankJ.1995Oligopoly in Loan and Deposit Markets: An Econometric Application to the NetherlandsDe EconomistVol. 143No. 3 pp. 353-66.

    • Search Google Scholar
    • Export Citation

    Vander VennetR.1997Determinants of EU Bank Takeovers: a Logit Analysismanuscript.

    VivesX.1999Banking Supervision in the European Monetary UnionmanuscriptInstitute d’Anàlisi Econòmica, Universitat Automona de BarcelonaBarcelono, Spain.

    • Search Google Scholar
    • Export Citation
1Joaquim Levy is currently on leave at the Brazilian Ministry of Finance. We would like to thank Tomas Baliño, Michael Deppler, Claudia Dziobek, Gianni de Nicoló, Garry Schinasi, Max Watson, participants at seminars at the IMF, Banque de France, Deutsche Bundesbank, Banco de España, BIS, ECB, and the authorities of the euro area countries for comments. Sergei Antoshin provided excellent research assistance. All remaining errors are our own.
2For the purpose of the paper, the euro-area countries are Austria, Belgium, Finland, France, Germany, Ireland, Italy, Luxembourg, the Netherlands, Portugal, and Spain.
3The paper focuses on the EU-11 (the euro area) countries; comparisons are made with the United Kingdom, the United States, and Japan, as appropriate. In the rest of the paper, the analyses should be understood as referring to this region, unless otherwise noted.
4This trend is examined in more detail in Subsection C below.
5In Spain, the large number of branches is attributable to a low population density, requiring many very small branches and ATMs to service bank customers. This is supported by data showing that the number of employees per branch is the lowest in Spain (Table 4).
7See below on profitability for more detail.
8This groups the savings banks under their parent organization, the Caisse Nationale des Caisses d’Epargne.
9The lack of legislation for the incorporation of “European” companies led to the initial design of the Dexia merger as a dual-listed company (in order to avoid the delicate issue of the nationality of the new company), in which two holding companies, Dexia France and Dexia Belgium, exchanged 50 percent of shares of the former CCB and CLF. In 1999 Dexia became a single holding company incorporated in Belgium and listed in Brussels, Luxembourg, and Paris.
10In Ireland, however, foreign banks have recently started to participate in domestic credit creation in the real estate market and have started to compete for deposits as well.
11The Netherlands has no savings banks, but has one large cooperative group.
12In a few special cases, these privileges are also provided to privately owned commercial banks to accomplish the same or similar public policy goals.
13The representativeness of the sample is discussed in detail in Annex I.
14Mutual fund assets in France, Italy, and Spain have come to represent between 40 percent and 80 percent of bank customers’ deposits in these countries, although they correspond to only about 10 percent of deposits in Germany.
15Information on mutual funds controlled by banks is generally disclosed independently of banks’ financial accounts. Although the income from asset management activities is an integral part of the profit and loss reports of banks, the portfolio of controlled mutual funds (including the holdings of liabilities issued by the associated institution) is typically not reported in banks’ balance sheets, or even as prominent off-balance sheet items. The separation of the balance sheets of banks and funds, although reflecting the fire walls existing within large financial groups and the fact that funds’ liabilities are fully collateralized, can nevertheless make banks’ consolidated accounts rather opaque and render an evaluation of the risk profile of financial institutions by the public or other counterparties difficult.
16Leveraged loans also command high fees and enjoy favorable treatment in terms of prudential capital required to cover their risks.
17The analysis of the magnitude of off-balance sheet items in banks’ reports sheds some light on the use of credit guarantees (e.g., in connection with securitization), but does not provide a clear view of the overall impact of disintermediation on the risk profile of banks. This is in part due to the lack of precision in the “credit risk equivalent” measure used to gauge risk of off-balance sheet items. With this in mind, the data available indicate that euro-area countries can be divided among (i) those in which the estimated risk of these banks’ items is high and exceeds 30 percent of banks’ total assets (France, Italy, Portugal, and Belgium), and (ii) those in which these items carry a relatively low risk level.
18The drop of the share of foreign assets relative to total assets in these countries did not imply that foreign assets held by banks decreased in absolute terms. As indicated by BIS data, Spain was among the euro-area countries for which exposures to foreign markets has grown the most in recent years.
19Although NPL are not published at an aggregate level and current laws do not require banks to disclose them in their accounting reports, some banks (notably the largest ones or those complying voluntarily with International Accounting Standards) do report these numbers. The banking supervisors receive these data on a regular basis, though it remains confidential.
20The transfer of bad loans from some large banks to special vehicles (“bad banks”) helped reduce the share of bad loans in total loans in both Italy (Banco di Napoli and Sicilcassa) and France (Crédit Lyonnais).
21It is important to note that, given the large share of public and nonincorporated banking institutions in many European countries as well as the differential accounting treatment of equity, banks’ returns on equity represent a less meaningful and reliable indicator of profitability than their returns on assets.
22The desire of banks’ core shareholders to retain control and ward off takeover attempts, while boosting Tier I capital, has led to the issuance of hybrid instruments (quasi-equity) combining some of the features of equity, and thus qualifying as capital, and some of the features of debt, such as the deductibility of interest payments and no voting powers. After some hesitation, regulators ultimately accepted this hybrid instrument as part of banks’ capital, although capping its contribution to Tier I capital.
23In Germany, part of banks’ “hidden reserves” are counted as assets in the BIS ratio, which explains some of the difference of the country’s equity-to-assets ratio with that in other countries. For tax reasons, banks have tended to largely understate them in their balance sheets. This discrepancy is likely to be reduced in the future since the German government’s 2000 tax reform includes a provision that eliminates the taxation of capital gains on equity holdings of corporate entities.
24Last year, the European Commission ruled that in Germany a Landesbank needed to repay its owners (the state (Land) and savings bank associations) interest on borrowed capital that had been obtained at substantially below market rates.
25ECB (1999b, 2000b).
26According to the BIS, imprudent lending is likely to worsen as global competition in the provision of financial services increases and management pays more attention to shareholder value. While it is possible that banks will respond by pricing risk more carefully, it is also possible that they will continue to be drawn into riskier ventures (BIS, 1999).
27In several countries, such a shift has been driven by the decline in interest rates, tax changes, and has also been favored by the buoyancy of stock markets in Europe and the elimination of exchange rate risk within the region.
28Efforts to avert this erosion through product differentiation are likely to have only a limited impact on the ultimate profitability of the industry.
29Last year, the European Commission felt obligated to intervene in a case deemed to have put national interests ahead of Community Law. It sent Portugal a reasoned opinion concerning measures taken by the Portuguese authorities to veto the acquisition of a controlling interest in the Champalimaud group by BSCH of Spain, stating “The Commission cannot allow defense of national interests by Member States to stand in the way of restructuring the EU’s financial services sector.” (European Commission, Financial Services, Legal Notice, October 20, 1999.)
30Soussa (1999) presents empirical evidence that too big to fail institutions enjoy lower funding costs and suggests that this leads to an uneven playing field, reducing competition and efficiency.
31The insurance industry in Europe is dominated by the Allianz (and its associate Münchener Rück), Axa, and Generali, all of which own equity stakes in the major European banks. The Dutch company, Aegon, and a few U.S. and U.K. companies share most of the rest of the market.
32See Foccarelli and others (1999), Berger, Demsetz, and Strahan (1999), and Houston, James, and Ryngaert (1999).
33Berger, Demsetz, and Strahan (1999).
34Deutsche Bank and Dresdner Bank, two large German banks, reportedly broke off merger talks after unresolvable differences regarding the fate of their respective investment banking units arose.
35Banco Bilbao Vizcaya Argentaria (Spain) and Unicredito (Italy) could not agree on a merged ownership structure, among other issues, because of the large difference in size.
36Domestic consolidation does not necessarily hurt competition: rivalry can be fierce even in a duopolistic structure. It does, however, increase the potential for cartel formation.
37Though defining the “market” is used colloquially, the more precise tool used in the United States to determine competitive effects examines the cross-elasticities of demand for the various products of the bank.
38See Vives (1999), OECD (1998). An important exception is Italy, where the central bank has aimed at protecting competition by mandating the closure or sale of branches and restrictions on the opening of new branches in several recent mergers, see Bank of Italy (1999).
40See Neven and von Ungern-Sternberg (1997).
42Swank (1995) estimated a model of financial sector behavior in the markets for mortgages and deposits during 1957-90 for the Netherlands and found evidence of oligopolistic behavior, though more recent evidence by the European Commission (1997) suggests that deposit spreads, a main indicator of collusive price behavior, have narrowed in euro-area countries (including the Netherlands) since the implementation of the 1992 internal market program.
43Collender and Shaffer (2000) show for a large sample of U.S. banks that out-of-market bank mergers or acquisitions need not, ceteris paribus, impair local economic growth, and may even have beneficial effects in rural markets.
44The development of such services is likely to depend on a greater penetration of simple electronic devices in households—instead of reliance on relatively expensive personal computers. This trend, foreshadowed in some ways in France with the introduction of the “minitel” in the 1980s, has been gathering pace with respect to other equipment, including portable telephones.
45In addition to the high marketing costs of any internet business, online banking has the added cost of the heavy investments in security systems necessary to gain the confidence of customers.
46The expansion of the French group Crédit Agricole in several other countries (notably Italy), and the purchase of a large stake in Deutsche Bank by La Caixa, a Spanish caja, are examples of such trend. On the other hand, their ownership structure may turn out to be a disadvantage for savings and mutual banks, since, as nonincorporated banks, they cannot easily engage in merger operations involving exchanges of shares with incorporated entities, and typically have to pay for their acquisitions in cash (for these reasons, Crédit Agricole is planning to create a holding company which would control an incorporated bank that would receive the assets of the bank and issue shares to the public).
47ABN-Amro and HSBC have reportedly, for instance, ruled out a takeover of Dresdner Bank on the grounds that expansion of its retail activities would be difficult, given the retail market structure in Germany. (“European Banks Rule Out German Takeover Bids,” Financial Times, April 12, 2000).
48In fact, the ongoing discussions at the Basle Committee toward a new capital adequacy framework point to market discipline as one of the three main pillars of effective oversight. Some of the larger nonincorporated institutions benefit from outside ratings of their short-term debt securities which may provide some market discipline, but without the ability to alter the management and control of the institution the degree to which such forces alter internal decision-making is limited.
49In the case in which publicly-traded banks suffer a principal-agent conflict, an unincorporated bank’s governance structure may be superior in so far as managerial benefits align closely to profit-maximization, when such profits are retained.
50Crockett (1997) highlights that the size of interbank exposures in the payments system has led to some observers to conclude that a disruption transmitted through the payments system is the largest single threat to financial stability.
51A major reason behind the predominance of unsecured transactions is the lagging integration of the repo market structures. Integration is hindered by differences in legal documentation, differences in acceptable collateral, and the lack of common rules for hypothecation of collateral across national borders.
52Overnight trade grew by close to 40 percent from late 1998 to mid 1999, and unsecured cross-border transactions increased by 130 percent over the same period (ECB 2000b and ECB 2000c). Though the proportion of unsecured transactions is not higher than in the United States or the United Kingdom, transactions among banks domiciled within these countries have common bankruptcy laws and legal standards. Bankruptcy codes are not uniform across the euro area.
53Real time gross settlement (RTGS) payments systems play a role in reducing risks by reducing the need for correspondent banking relationships and lowering the potential transmission of shocks due to liquidity problems at the end of the day. In this respect, TARGET, a collateralized RTGS payment system for the euro area, has sped up electronic transfers across borders and has encouraged economies of scale in back-office operations. Large banks, with other options for settlement, have chosen TARGET for their very high-value payments as it offers advantages in terms of liquidity management (ECB, 1999b).
54Cross border interbank transactions have become increasingly important, exceeding euro 85 billion a day and accounting now for more than 50 percent of the interbank market. The 20 largest banks account for 40 percent of total interbank assets and liabilities in the euro area (ECB, 2000c).
55Berger, Demsetz, and Strahan (1999).
57Some policymakers have argued that supervisors are not likely to be able to keep up with new developments and thus policies should be put in place that make the activities of these big institutions more transparent to other market participants (e.g., through the use of increased financial disclosure, published credit evaluations, information about flows of various instruments through the markets) with the goal of permitting market forces to play a greater role in stemming poor behavior on the part of an institution. Other policymakers are skeptical about whether financial markets can be relied upon to appropriately penalize institutions in a timely fashion for even obvious risks. For an in-depth discussion on the implications of conglomeration for the financial safety net, see Saapar and Soussa (1999).
58The risk inherent to e-commerce business—in addition to that arising from the uncertain prospects of financial returns to the investments required to initiate these activities in large scale—is in part that the reputation of the bank may be affected by problems with the security or timeliness of payments. Also, risks are altered in the cases in which the capital of the bank entering into e-commerce is tied to that of the telecom or the Internet Service Provider (ISP) companies (e.g., through cross shareholdings).
59The experience in several European countries has been that these schemes were effective in protecting depositors in small banks (e.g., the French Pallas Stern and the Italian Sicilcassa), but government help was necessary in the case of large banks (e.g., Credit Lyonnais, Banco di Napoli and, to an extent, Banesto), especially before the national schemes had fully complied with the EU directives. Currently, governments’ scope to provide risk capital may be limited by the European Commission’s limit on state aid.
60This is a purely demonstrative metric that may be considered as a lower bound. In many bank failures, the final losses amounted to several times the capital at the time of the initial distress.
61“International Capital Markets: Developments, Prospects, and Key Issues,” (1999), p. 135.
62The U.S. Congress has also set up a special mechanism within the new Gramm-Leach-Bliley Act for the oversight of financial holding companies—including foreign-owned ones. This scheme, while relying on the cooperation among the array of functional supervisors that characterizes the U.S. supervisory institutions, aims at enabling the Federal Reserve Board to acquire an intimate and frequently updated knowledge of the overall risk profile, strategy, and policies of the top banks. See Federal Reserve System Board of Governors (1999).
63The Joint Forum is comprised of an equal number of senior bank, insurance, and securities supervisors representing each supervisory constituency. Thirteen countries are represented in the Joint Forum: Australia, Belgium, Canada, France, Germany, Italy, Japan, the Netherlands, Spain, Sweden, Switzerland, the United Kingdom and the United States. The EU Commission attends in an observer capacity.
64For information, see the Basel Committee’s work in financial conglomerates on http:\\
65“Report on Financial Stability,” EU Economic and Financial Committee, (2000).
66In practice, the ability to distinguish between liquidity and solvency problems has been a difficult one and thus it is important not to permit gaps in the processes for detecting their development.
67See Prati and Schinasi (1999) for a more detailed discussion.
68See the example in Baliño and Ubide (2000). Consider a bank domiciled in country A whose operations in country B, although small relative to the global operations of the bank, make it one of the dominant banks in country B. If that bank was to become insolvent, would the bank’s home country supervisor internalize the massive disruption of activity in country B that would result from the bank’s liquidation and thus be willing to provide financial support—which would entail a wealth transfer from country A to country B?
69In Germany, for instance, mortgage banks are licensed and supervised under a different set of rules than other banks.
70As an example of transferring supervision following a significant change in the geographical scope of operations, the supervision of the Spanish cajas was originally a responsibility of regions, which—with the expansion of the cajas outside their home regions—was delegated to the Bank of Spain.
72In fact, as long as the financial responsibility in case of a bank failure remains with the national authorities, the decision on the closure of a bank should remain with the national supervisor to avoid incentive problems.
73In this respect, the example of arrangements in the United Kingdom is illuminating. Supervision is performed by the Financial Services Authority (FSA). However, the Bank of England continues to analyze the balance sheet of some individual institutions that are considered to be important for the stability of the financial system, with an emphasis on system wide vulnerabilities rather than on the conventional prudential analysis performed by the FSA (see Hawkesby (1999)).

Other Resources Citing This Publication