A. Overview

177. Financial integration promises a cheaper and better allocation of resources and risk, increased liquidity, and greater financial stability. The fragmentation of many banking markets dates back to public policies adopted in response to the financial crises of the 1930s. Following the crises, the consensus was that restraining competition would help in preserving the stability of the banking and financial industry. Accordingly, banks were often forbidden to operate across states or regions or to offer a full array of financial services. The allocation of credit was controlled and credit ceilings played a key role in monetary policy. However, attitudes have since evolved, partly because of developments on the ground that were driven by profit maximization and technological progress. Policymakers and supervisors now emphasize the benefits of competition and market discipline, supported by capital adequacy regulation and close supervision; and the academic literature agrees that there is no simple trade-off between competition and financial stability.¹³³

178. This paper reviews the EU and US histories of financial integration and analyzes the current state of play through the prism of efficiency and competition. Section B presents the milestones of financial integration in Europe and the United States and briefly reviews remaining obstacles. Section C explores the differences between the roles of banks and markets in Europe and the United States. The main argument is that Europe’s financial markets are less complete and thus the division of tasks between money and capital markets on the one hand and banks on the other may not have gone as far as in the United States. Section D presents evidence suggesting that EU banks are less effective in raising revenue than their US counterparts, with differences in their business mix and specialization potentially playing an important role. Section E explores competition, finding that larger, internationally-active banks engage in more competition in Europe than smaller banks; also, European banks appear to engage in less competition than their US counterparts. Section F reflects further on the reasons for the observed, lower revenue effectiveness of EU banks, suggesting that less competition, missing financial markets, and various other factors might play a role. Section G discusses policy implications.

179. While US financial markets offer a natural benchmark, they do not necessarily represent a model for convergence. Financial markets in the United States offer a natural benchmark for assessing integration in the European Union partly because of the country’s size and level of economic development, but also because banking markets were highly fragmented in both areas until the drive for integration accelerated in the late 1980s. Money and capital markets, by contrast, as well as regulation and supervision were always highly integrated in the United States but not in the European Union. This is likely to have influenced the relative developments of banks and markets, a theme that is revisited below. One caveat is that financial market structures differ quite considerably across countries in Europe. In that sense, by focusing on Europe as a whole, various complexities of the area’s financial sector are not discussed. Instead, whenever particularly relevant, this paper draws on the existing evidence on country specificities.¹³⁴

B. History of Integration

180. Through the late 1980s, capital and banking markets in the European Union were highly fragmented but this changed quickly with the quest for a single currency. Aside for the Treaty of Rome, key milestones for European financial integration include, the Directive on Liberalization of Capital flows (1988), the Second Banking Directive (1989), and the Maastricht treaty (1992). The Maastricht treaty set the stage for the single currency and the eventual integration of bank, money and capital markets but left regulatory and supervisory powers with national authorities. The 1988 directive opened capital flows effective July 1990 but allowed the reimposition of controls under emergency circumstances. The Second Banking Directive entered into force starting in 1993 and set out the key drivers of banking market integration: home-country control and mutual recognition, resulting in a “single passport”. Any bank licensed in an EU country was subsequently free to open branches in any other EU country provided it met some common, minimum standards.

181. Banking market integration in the United States was a more gradual bottom-up process, taking place against the backdrop of a single currency and integrated money and capital markets. The federal law that prohibited commercial banks from operating across state lines dated back to 1927: the McFadden Act did, however, permit cross-border banking through multibank holding companies with state approval. In 1978, Maine took the first step and allowed entry of bank holding companies from other states, provided these states reciprocated. By 1992, virtually all states had passed reciprocal entry laws of some sort.¹³⁵ The Riegle-Neal Interstate Banking and Branching Efficiency Act capped this development, allowing national bank branches across state lines after June 1, 1997. As a result, between the mid-1970s and mid-1990s, the ratio of a typical state bank’s assets held by an out-of-state bank holding company climbed from 10 to over 60 percent.¹³⁶ In addition, the Gramm-Leach-Bliley Act of 1999 ended the separation between commercial and investment banking that dated back to the Glass-Steagall Act of 1933.

182. Reciprocity and the mutual recognition of standards were key drivers of financial integration in both Europe and the United States. In the United States, this was, to some extent, driven by loopholes in federal legislation and fostered by the highly integrated legislative and regulatory frameworks. In Europe, mutual recognition offered a much quicker vehicle for integration than the harmonization of laws and regulations. Nonetheless, a minimum of harmonization is required for an integrated financial market to emerge. Also, the mutual recognition extends only to branches and not subsidiaries: the latter avenue for entering foreign markets remains more costly to pursue. More generally, the process of integration in Europe was multilateral and part of a wider, top-down program to achieve a single market. As a result, ownership of some EU countries might not have been as strong as that of single US states entering “bilateral” agreements. But ultimately financial integration will have to ensure the free flow of services across EU member countries foreseen by the Treaty of Rome.

183. Notwithstanding significant progress, many observers thus come to a mixed assessment of what has been achieved in Europe.¹³⁷ Integration has proceeded furthest at the wholesale level, while the market for retail services is lagging behind.¹³⁸ Also, regulation and supervision continue to differ significantly.

• Regarding financial markets, wholesale money and bond markets are now relatively well integrated. However, a fully satisfactory degree of integration has only been achieved in the unsecured euro money market. The markets for corporate bonds and commercial paper are expanding rapidly. But other markets, such as those for asset-backed securities, lag far behind. Equity markets are converging, as evidenced by less home bias and increasing correlations of returns but the process is incomplete. However, many small exchanges continue to operate, even within countries. Crucially, crossborder clearing and settlement are far from integrated.

• Banking remains fragmented. The Single Passport has fostered crossborder branching: the market share of crossborder branches in the EU reached some 24 percent of GDP in 2002. But the attractive avenue of entering markets through subsidiaries is complex. Relatedly, crossborder mergers and acquisitions (M&A) face significant hurdles and thus have not featured prominently: during 1987–2003, they accounted for less than 10 percent of all bank M&A activity in 13 years out of the 17 years in the sample and reached peaks of close to 30 percent only in 1987 and 2000.¹³⁹ Also, only some 5 percent of bank credit is granted across borders and the variety of mortgage products, for example, is presently rather limited in many EU countries.¹⁴⁰ However, retail banking will probably retain a strong geographic component. Even in the United States no bank presently operates a dense branch network across the entire country. While some 538 organizations operated branches in more than one state as of mid-2003, only fourteen had branches in more than ten states. The institutions with the widest geographic reach had branches in only about half the states.¹⁴¹

• Regulation and supervision of banks, insurance companies, and securities markets still differ significantly across EU countries, slowing the pace of integration. While considerable progress has been made with the FSAP and the Lamfalussy process, a single rules book and uniform supervisory practices are still a long way off.

• Institutional differences continue to hamper the development of a unified market for financial services. Crossborder barriers derive from tax legislation, for example, from double taxation of income flows of associated companies established in different countries; or from the taxation of savings, including tax breaks and other vehicles to support personal savings, notably for retirement. In addition financial reporting standards differ across countries for non-listed companies; European private law is not sufficiently consolidated, making it difficult to arrange crossborder collateral pledges; and different consumer protection regimes stand in the way of the introduction of EU-wide retail banking products.¹⁴²

C. Structural Characteristics

Banks and capital markets

184. Bank- and market-based systems offer different advantages. An extensive literature discusses the relative merits of each system without reaching definitive conclusions.¹⁴³ One key argument in favor of banks is that their long-term relationship with firms helps overcome the inefficiencies related to adverse selection and moral hazard (Stiglitz and Weiss, 1983). Hence banks may better smooth intertemporal risk than markets. However, solid empirical evidence on these intertemporal risk sharing properties is lacking and countries have a variety of public programs that are likely to share risk more effectively across time (for example, defined benefit pension systems). Market-based systems, by contrast, are viewed as providing better cross-sectional risk sharing. Also, such systems generate more information and thus are seen by some to better fit advanced economies, as these economies explore new production possibilities rather than catch up with existing frontiers (e.g., Boot and Thakor, 1997). However, markets are very volatile and require a complex legal and regulatory infrastructure to work well.

185. Relative to the economy, EU capital markets are much smaller and bank balance sheets much larger than in the United States(Table VI.1). The absence of a single currency and legal/regulatory hurdles in the European Union contributed to the area’s smaller capital markets. Thus, the EU financial system is typically labeled bank-based and the US system market-based.

Table VI.1.

Euro-Area and United States: Banks and Markets, 2004

(In percent of nominal GDP)

Sources: US FED, ECB, and European Securitization Forum.

^1/For euro area, including Eurosystem. For US, commercial banks only.

^2/From consolidated balance sheet of euro-area MFIs.

Table VI.1.

Euro-Area and United States: Banks and Markets, 2004

(In percent of nominal GDP)

		Euro area	United States
Bonds			123	149
	Government		58	37
	Banks 1/2/		48	6
	Non-bank financial institutions		10	27
	Non-financial corporations		8	25
	Agencies		0	53
	Other		0	0
Equity			53	147
Bank assets 1/			208	92
	Bank loans to nonfinancial private sector		92	118
		Firms	42	30
		Households	50	87
	Bank loans to general government		11	0
	Bank loans to non-bank financial corporates		8	...
Memorandum item:
Asset-backed securities (stock)			...	59
Asset-backed securities (new issuance)			4	...

Sources: US FED, ECB, and European Securitization Forum.

^1/For euro area, including Eurosystem. For US, commercial banks only.

^2/From consolidated balance sheet of euro-area MFIs.

View Table

Table VI.1.

Euro-Area and United States: Banks and Markets, 2004

(In percent of nominal GDP)

		Euro area	United States
Bonds			123	149
	Government		58	37
	Banks 1/2/		48	6
	Non-bank financial institutions		10	27
	Non-financial corporations		8	25
	Agencies		0	53
	Other		0	0
Equity			53	147
Bank assets 1/			208	92
	Bank loans to nonfinancial private sector		92	118
		Firms	42	30
		Households	50	87
	Bank loans to general government		11	0
	Bank loans to non-bank financial corporates		8	...
Memorandum item:
Asset-backed securities (stock)			...	59
Asset-backed securities (new issuance)			4	...

Sources: US FED, ECB, and European Securitization Forum.

^1/For euro area, including Eurosystem. For US, commercial banks only.

^2/From consolidated balance sheet of euro-area MFIs.

186. It is not clear, however, whether the distinction between bank- and market-based systems is very relevant for an EU-US comparison. For example, while EU banks hold larger balance sheets, the data point to less bank lending to the nonfinancial private sector by EU banks rather than US banks. EU banks only hold a relatively small lead with respect to lending to firms. Because of securitization––which took off during the 1980s with the development of collateralized mortgage obligations––loans to the nonfinancial private sector in the United States do not necessarily show up on banks’ balance sheets as assets (Table VI.1). Specifically, US banks no longer fund directly much of their household lending (home mortgages and, more recently, consumer credit), which is well suited to standardization and thus securitization. Since US banks also face rising competition from bond markets for unsecuritized business lending, they increasingly hold loans that are less well suited to standardization, notably business loans collateralized by real estate.¹⁴⁴

Accordingly, US banks increasingly engage in placing risk in financial markets, keeping only those risks on balance sheets for which they enjoy a particular comparative advantage. Notwithstanding the large difference with respect to balance sheet size, EU and US per capita banking sector employment is thus quite similar (Table VI.2). From that perspective, banks are similarly important in both economies. Where Europe leads is in the accumulation of bricks and mortar (branches) and where it lags is in equity markets.

Table VI.2.

EU 15 and United States: Structural Indicators, 1997-2003

Sources: ECB, FDIC, WEO database, and Pilloff (2004).

^1/In full-time equivalents; raw numbers are likely to be higher.

^2/The data here have a slightly different coverage from those in Table 1.

Table VI.2.

EU 15 and United States: Structural Indicators, 1997-2003

			1997	1998	1999	2000	2001	2002	2003
Number of banks		US	10,923	10,464	10,223	9,904	9,615	9,354	9,182
		EU 15	9,624	9,337	8,872	8,433	8,084	7,751	7,444
	(Per million people)	US	41	39	37	36	34	32	32
		EU 15	26	25	24	22	21	20	19
Number of bank branches		US	73,366	74,798	76,576	76,567	78,123	78,503	79,756
		EU 15	202,092	200,319	198,973	197,513	194,349	190,279	186,009
	(Per million people)	US	273	276	280	278	274	272	274
		EU 15	539	534	529	523	512	500	487
Number of bank employees		US1/	1,784,482	1,863,910	1,901,596	1,914,517	1,967,622	2,017,645	2,045,976
		EU 15	...	...	2,820,563	2,821,845	2,862,338	2,823,784	2,774,175
	(Per million people)	US	6,651	6,883	6,960	6,946	6,889	6,994	7,023
		EU 15			7,493	7,467	7,543	7,417	7,264
Banking system assets2/		US	73	75	74	76	78	80	82
	(In percent of GDP)	EU 15	184	185	195	203	213	211	214

Sources: ECB, FDIC, WEO database, and Pilloff (2004).

^1/In full-time equivalents; raw numbers are likely to be higher.

^2/The data here have a slightly different coverage from those in Table 1.

View Table

Table VI.2.

EU 15 and United States: Structural Indicators, 1997-2003

			1997	1998	1999	2000	2001	2002	2003
Number of banks		US	10,923	10,464	10,223	9,904	9,615	9,354	9,182
		EU 15	9,624	9,337	8,872	8,433	8,084	7,751	7,444
	(Per million people)	US	41	39	37	36	34	32	32
		EU 15	26	25	24	22	21	20	19
Number of bank branches		US	73,366	74,798	76,576	76,567	78,123	78,503	79,756
		EU 15	202,092	200,319	198,973	197,513	194,349	190,279	186,009
	(Per million people)	US	273	276	280	278	274	272	274
		EU 15	539	534	529	523	512	500	487
Number of bank employees		US1/	1,784,482	1,863,910	1,901,596	1,914,517	1,967,622	2,017,645	2,045,976
		EU 15	...	...	2,820,563	2,821,845	2,862,338	2,823,784	2,774,175
	(Per million people)	US	6,651	6,883	6,960	6,946	6,889	6,994	7,023
		EU 15			7,493	7,467	7,543	7,417	7,264
Banking system assets2/		US	73	75	74	76	78	80	82
	(In percent of GDP)	EU 15	184	185	195	203	213	211	214

Sources: ECB, FDIC, WEO database, and Pilloff (2004).

^1/In full-time equivalents; raw numbers are likely to be higher.

^2/The data here have a slightly different coverage from those in Table 1.

187. The data on employment and lending put a different spin on the public debate about consolidation.¹⁴⁵ The European Union is often viewed as overbanked and in need of consolidation––not least owing to the large number of credit institutions in Germany, France, and Italy––amid accelerating disintermediation spurred by growing money and capital markets. But, the per capita number of banks and banking sector employment in the United States are similar or higher than in the European Union. Furthermore, although the number of banks has been falling significantly, total banking sector employment has actually risen in the United States over the past five years.¹⁴⁶ Over a horizon spanning the past three decades, the number of banks was virtually halved in the United States. However, the number of branches doubled and continues to grow.

Bank performance

188. EU banks do not display better financial strength indicators than their US counterparts (Table VI.3). Data for 2003 suggest that the pre-tax return on assets (ROA) of EU banks reached only one third of the level of that of US banks. Reflecting a greater importance of off-balance sheet activities for US banks the gap with respect to return on equity (ROE) is considerably smaller but still quite significant. While the cost ratios of EU banks are lower, the revenue ratios fall short of those of US banks by an even wider margin, with interest and other revenues contributing in similar proportions. Cross sectional data on the top 100 EU and US banks paint the same picture (Table VI.4): not only does the median bank appear less profitable and capitalized in the EU, the same holds for the 10 percent weakest banks in the sample. Also, these data suggest that the relation between size on the one hand and financial strength indicators on the other is unclear, as evidenced by a comparison of larger (top 50) and smaller (lower 50) banks in the sample.¹⁴⁷

Table VI.3.

EU 15 and United States: Indicators of Bank Profitability and Efficiency, 2003

(In percent of assets, unless otherwise noted)

Sources: ECB (EU Banking Sector Stability, 2004) and FDIC (Quarterly Banking Profile, 2003).

Table VI.3.

EU 15 and United States: Indicators of Bank Profitability and Efficiency, 2003

(In percent of assets, unless otherwise noted)

		EU 15	US
Net interest income		1.4	3.3
Net non-interest income		1.0	2.5
Total income		2.4	5.8
Staff costs		0.9	1.4
Other costs		0.6	1.9
Total costs		1.4	3.3
Operating profits		0.9	2.5
Profits, pre-tax		0.6	2.1
Profits, after tax		0.4	1.4
Return on equity
	(In percent of Tier 1)	9.9	15.3
Equity-to-asset ratio		4.2	9.2
Risk-based Tier 1 ratio (In percent)		8.8	10.1
Overall solvency ratio (In percent)		12.4	12.7

Sources: ECB (EU Banking Sector Stability, 2004) and FDIC (Quarterly Banking Profile, 2003).

View Table

Table VI.3.

EU 15 and United States: Indicators of Bank Profitability and Efficiency, 2003

(In percent of assets, unless otherwise noted)

		EU 15	US
Net interest income		1.4	3.3
Net non-interest income		1.0	2.5
Total income		2.4	5.8
Staff costs		0.9	1.4
Other costs		0.6	1.9
Total costs		1.4	3.3
Operating profits		0.9	2.5
Profits, pre-tax		0.6	2.1
Profits, after tax		0.4	1.4
Return on equity
	(In percent of Tier 1)	9.9	15.3
Equity-to-asset ratio		4.2	9.2
Risk-based Tier 1 ratio (In percent)		8.8	10.1
Overall solvency ratio (In percent)		12.4	12.7

Sources: ECB (EU Banking Sector Stability, 2004) and FDIC (Quarterly Banking Profile, 2003).

Table VI.4.

EU 15 and US Banking Sector Indicators, 1997-2003

(In percent, unless otherwise noted)

Source: Fitch IBCA database; and IMF staff calculations.

^1/In percent of assets.

^2/Operating revenue excluding interest expenses.

^3/Operating costs plus interest expenses.

^4/Sum of average return on assets and Tier 1 ratio divided by variance of average return on assets.

Table VI.4.

EU 15 and US Banking Sector Indicators, 1997-2003

(In percent, unless otherwise noted)

		EU banks			US banks
		All	Small	Large	All	Small	Large
Return on average assets
	10th percentile	0.08	0.05	0.10	0.40	0.44	0.37
	Median	0.46	0.42	0.47	1.22	1.26	1.18
Return on average equity
	10th percentile	2.64	1.94	3.74	7.10	7.86	6.82
	Median	11.42	11.10	11.61	14.84	14.18	15.43
Tier 1 ratio
	10th percentile	5.60	5.68	5.52	7.30	7.70	6.90
	Median	7.40	7.50	7.40	9.45	9.80	8.50
Equity ratio
	10th percentile	2.01	1.79	2.41	4.73	5.78	4.09
	Median	3.87	4.10	3.80	8.14	8.51	7.66
Revenue 1/2/
	10th percentile	4.19	4.07	4.33	5.15	5.38	4.84
	Median	5.92	5.67	6.08	8.07	8.17	7.92
Cost 1/3/
	90th percentile	7.57	7.34	7.65	10.40	10.49	9.94
	Median	5.30	5.19	5.52	6.37	6.42	6.28
Operating profit 1/
	10th percentile	0.09	0.07	0.11	0.22	0.22	0.22
	Median	0.56	0.51	0.60	1.25	1.46	1.06
Operating expenditure 1/
	90th percentile	3.03	3.34	2.86	5.30	5.82	4.41
	Median	1.69	1.58	1.75	2.67	2.73	2.48
Personell expenditure 1/
	90th percentile	1.40	1.46	1.34	2.16	2.05	2.18
	Median	0.78	0.69	0.84	1.31	1.31	1.32
Other revenue 1/
	Median	1.04	0.99	1.11	1.59	1.42	1.80

Source: Fitch IBCA database; and IMF staff calculations.

^1/In percent of assets.

^2/Operating revenue excluding interest expenses.

^3/Operating costs plus interest expenses.

^4/Sum of average return on assets and Tier 1 ratio divided by variance of average return on assets.

View Table

Table VI.4.

EU 15 and US Banking Sector Indicators, 1997-2003

(In percent, unless otherwise noted)

		EU banks			US banks
		All	Small	Large	All	Small	Large
Return on average assets
	10th percentile	0.08	0.05	0.10	0.40	0.44	0.37
	Median	0.46	0.42	0.47	1.22	1.26	1.18
Return on average equity
	10th percentile	2.64	1.94	3.74	7.10	7.86	6.82
	Median	11.42	11.10	11.61	14.84	14.18	15.43
Tier 1 ratio
	10th percentile	5.60	5.68	5.52	7.30	7.70	6.90
	Median	7.40	7.50	7.40	9.45	9.80	8.50
Equity ratio
	10th percentile	2.01	1.79	2.41	4.73	5.78	4.09
	Median	3.87	4.10	3.80	8.14	8.51	7.66
Revenue 1/2/
	10th percentile	4.19	4.07	4.33	5.15	5.38	4.84
	Median	5.92	5.67	6.08	8.07	8.17	7.92
Cost 1/3/
	90th percentile	7.57	7.34	7.65	10.40	10.49	9.94
	Median	5.30	5.19	5.52	6.37	6.42	6.28
Operating profit 1/
	10th percentile	0.09	0.07	0.11	0.22	0.22	0.22
	Median	0.56	0.51	0.60	1.25	1.46	1.06
Operating expenditure 1/
	90th percentile	3.03	3.34	2.86	5.30	5.82	4.41
	Median	1.69	1.58	1.75	2.67	2.73	2.48
Personell expenditure 1/
	90th percentile	1.40	1.46	1.34	2.16	2.05	2.18
	Median	0.78	0.69	0.84	1.31	1.31	1.32
Other revenue 1/
	Median	1.04	0.99	1.11	1.59	1.42	1.80

Source: Fitch IBCA database; and IMF staff calculations.

^1/In percent of assets.

^2/Operating revenue excluding interest expenses.

^3/Operating costs plus interest expenses.

^4/Sum of average return on assets and Tier 1 ratio divided by variance of average return on assets.

189. EU banks appear to engage in less risky activities than US banks and hold somewhat lower risk-adjusted capital. This can be gleaned from the relation between the simple equity-to-asset ratio and the risk-based capital ratios. The simple equity-to-asset ratio of US banks is more than twice as high as that of EU banks but not the regulatory solvency ratio (Table VI.3). In other words, for regulatory purposes the assets of EU banks carry lower risk weights, suggesting that they engage in less risky activities. By contrast, capitalization of US banks has reached the highest level in some 50 years, owing to greater risk exposures and the market’s increased demand that banks’ default risk be adequately priced (Flannery and Rangan, 2004). The high level of capital cuts the return on equity (ROE) of US banks relative to that of EU banks but still leaves it some 50 percent higher.

190. The gap of EU relative to US financial strength indicators opened in the 1990s and appears to reflect a trend rather than cyclical development (Figure VI.1). Several factors might explain the improved performance of US banks, including market exit of weaker players,¹⁴⁸ cross-state M&A activity, and the accelerated development of new financial markets, including securitization, that enabled banks to better leverage their comparative advantages. By contrast, over the past decade the financial strength indicators for the European Union have moved broadly sideways. Section D sheds further light on the performance differences.

EU and US: Profitability and the Economic Cycle

(In percent, unless otherwise noted)

Citation: IMF Staff Country Reports 2005, 266; 10.5089/9781451813029.002.A006

Sources: ECB; OECD, Bank Profitability, 2002; and Federal Deposit Insurance Corporation.Note: Data for the European Union from 1988 to 2001 refer only to Germany, France, Italy, UK, and Spain.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

EU and US: Profitability and the Economic Cycle

(In percent, unless otherwise noted)

Citation: IMF Staff Country Reports 2005, 266; 10.5089/9781451813029.002.A006

Sources: ECB; OECD, Bank Profitability, 2002; and Federal Deposit Insurance Corporation.Note: Data for the European Union from 1988 to 2001 refer only to Germany, France, Italy, UK, and Spain.

• Download Figure
• Download figure as PowerPoint slide

EU and US: Profitability and the Economic Cycle

(In percent, unless otherwise noted)

Citation: IMF Staff Country Reports 2005, 266; 10.5089/9781451813029.002.A006

Sources: ECB; OECD, Bank Profitability, 2002; and Federal Deposit Insurance Corporation.Note: Data for the European Union from 1988 to 2001 refer only to Germany, France, Italy, UK, and Spain.

• Download Figure
• Download figure as PowerPoint slide

D. Productive Efficiency

191. The lower profitability of EU banks could be related to lower efficiency or other factors, including different business models with lower risks. Profitability of EU banks could have been lower because banks (for various reasons) face less pressure to use their inputs efficiently—this would be captured as a lower X-efficiency. Simple comparisons of profitability, revenue, and cost indicators do not provide enough information to judge the operational effectiveness of EU relative to US banks. For example, EU banks may, over the period under study, have faced higher labor costs and a less favorable yield curve than their US counterparts. Notwithstanding an efficient use of inputs, EU banks’ profits may have been lower as a result. In other words, to judge efficiency it is important to hold constant for different input costs, which requires estimating revenue and cost functions. Furthermore, differences in business models and risks need to be considered as well. This section follows an approach that has been widely used in the literature to estimate X-efficiency.¹⁴⁹

192. Gauging productive efficiency requires that an assumption be made about banks’ activities. According to the “intermediation approach,” which is followed here, banks intermediate financial services using labor and capital as inputs, with the values of loans and investments used as the output measure. Given that labor and capital are the inputs, operating costs plus interest expenses are the relevant cost measures. The relevant revenue measure is operating revenue, excluding interest expenses.¹⁵⁰

193. A stochastic “best practices” frontier is a useful tool to gauge banks’ efficiency. This approach specifies the functional form of the efficient frontier as a translog cost or revenue function to investigate, respectively, the effectiveness of cost control and revenue generation:¹⁵¹

with lower case letters denoting natural logarithms. Specifically, each bank i produces two outputs y (loans and other earning assets) and relies on three inputs with prices p (labor, interest expenses, and other operating costs). In addition, the equation includes a set of exogenous variables Z, two time dummies D (which proxy for changes in the macroeconomic environment), and a constant α. The dependent variable x denotes either operating revenue, excluding interest expenses or operating cost plus interest expenses for bank i in year t; and the dummy β_EU for European banks measures their relative management effectiveness on both accounts.

194. Importantly, an effort needs to be made to hold constant for differences in banks’ business models for a fair comparison of effectiveness. Section C shows that banks are similarly important in Europe and the United States. However, US banks increasingly keep only those assets on balance sheet for which they enjoy a particular comparative advantage (selling off other assets in markets), and these appear to be riskier ones. The set of exogenous variables Z in the regression equation tries to capture these differences in business models between banks. These variables comprise the loan-to-asset (L/A), deposit-to-liability (D/L), asset-to-employee (A/E), and the equity-to-asset ratios (C/A), depending on the specific regression. Several examples illustrate the role of these variables: (i) one bank may engage extensively in securitization of less risky assets and keep more risky assets on its balance sheet. Another engages in the same lending but keeps all assets on balance sheet. The former bank would have to hold more equity relative to assets than the latter and this would be captured by the equity-to-asset-ratio; (ii) one bank might rely relatively more on deposits as a source of funding than others that, instead, tap the bond market. This difference would be captured by the deposit-to-liability ratio; and (iii) one bank may be more active in investment banking and other services than others that, instead, focus on firm/household lending. This would be captured by a lower loan-to-asset ratio. Furthermore, the regressions distinguish between the top 50 and the lower 50 banks, as their size differs considerably and so might their business models in ways not captured by the Z variables. Nonetheless, the exogenous variables Z and the splitting of the sample clearly cannot proxy perfectly the differences in business models of banks, including their risks.¹⁵²

195. The EU sample of banks is more homogenous than the US sample and the EU banks tend to hold more assets. The data sample comprises the 100 largest banks in the European Union and the United States, respectively, for 1997, 2000, and 2003, drawn from Bankscope. It is difficult to put an exact number on the market share of these 100 banks in each area but it exceeds 50 percent, probably by a substantial margin. The combined assets of the top 50 banks (“large” banks) are about four (seven) times as large as those of the lower 50 banks (“small” banks) in the European Union (United States). While the EU-to-US ratio of median assets equals 3.2, the same ratio for median employment only reaches 1.5, again pointing to the greater role of off-balance sheet activities among US banks.

196. Ordinary least squares estimates of the efficient frontier suggest that EU banks are less effective in generating revenue, while costs appear well behaved.¹⁵³ On average, EU banks exhibit 5 percent lower costs than their US counterparts, regardless of the regression specification (Table VI.5). However, they also generate up to about 18 percent less revenue, with the gap falling to about 12 percent upon including the capital-to-asset ratio in the regression (Table VI.6).¹⁵⁴ Recall that US banks feature considerably higher standard capital ratios than EU banks, while their regulatory capital ratios are broadly comparable. This is because they engage in more off-balance sheet activities and hold riskier assets. The capital-to-asset ratio proxies for this difference in asset-mix and thus reduces the revenue effectiveness gap of EU banks. Further differentiating between the larger and the smaller banks cuts the revenue efficiency gap of EU banks to some 7 percent, while the cost advantage falls to some 3 percent.

Table VI.5.

Measures of Cost Efficiency of EU Banks (Relative to US banks), 1997-2003

(Dependent variable is the log of operating expense plus interest expense)

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

^***denotes that the coefficient is significantly different from zero at the 1 percent level.

^**denotes that the coefficient is significantly different from zero at the 5 percent level.

^*denotes that the coefficient is significantly different from zero at the 10 percent level.

Table VI.5.

Measures of Cost Efficiency of EU Banks (Relative to US banks), 1997-2003

(Dependent variable is the log of operating expense plus interest expense)

Equation	Efficiency coefficient	Rbar2	Degrees of freedom
All banks
No exogenous variable	-4.81	***	0.99	423
L/A, D/L, A/E ratios added	-4.66	***	0.99	414
L/A, D/L, A/E, C/A ratios added	-5.21	***	0.99	413
Large banks
No exogenous variable	-4.01	***	0.99	182
L/A, D/L, A/E ratios added	-4.37	***	0.99	174
L/A, D/L, A/E, C/A ratios added	-3.99	**	0.99	173
Small banks
No exogenous variable	-2.13		0.99	223
L/A, D/L, A/E ratios added	-2.61		0.99	219
L/A, D/L, A/E, C/A ratios added	-2.64		0.99	218

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

^***denotes that the coefficient is significantly different from zero at the 1 percent level.

^**denotes that the coefficient is significantly different from zero at the 5 percent level.

^*denotes that the coefficient is significantly different from zero at the 10 percent level.

View Table

Table VI.5.

Measures of Cost Efficiency of EU Banks (Relative to US banks), 1997-2003

(Dependent variable is the log of operating expense plus interest expense)

Equation	Efficiency coefficient	Rbar2	Degrees of freedom
All banks
No exogenous variable	-4.81	***	0.99	423
L/A, D/L, A/E ratios added	-4.66	***	0.99	414
L/A, D/L, A/E, C/A ratios added	-5.21	***	0.99	413
Large banks
No exogenous variable	-4.01	***	0.99	182
L/A, D/L, A/E ratios added	-4.37	***	0.99	174
L/A, D/L, A/E, C/A ratios added	-3.99	**	0.99	173
Small banks
No exogenous variable	-2.13		0.99	223
L/A, D/L, A/E ratios added	-2.61		0.99	219
L/A, D/L, A/E, C/A ratios added	-2.64		0.99	218

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

^***denotes that the coefficient is significantly different from zero at the 1 percent level.

^**denotes that the coefficient is significantly different from zero at the 5 percent level.

^*denotes that the coefficient is significantly different from zero at the 10 percent level.

Table VI.6.

Measures of Revenue Efficiency of EU Banks (Relative to US banks), 1997-2003

(Dependent variable is the log of operating income plus interest expense)

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.1/ Confidence region is two standard errors wide on each side of point estimate.

^***denotes that the coefficient is significantly different from zero at the 1 percent level.

** denotes that the coefficient is significantly different from zero at the 5 percent level.

^*denotes that the coefficient is significantly different from zero at the 10 percent level.

Table VI.6.

Measures of Revenue Efficiency of EU Banks (Relative to US banks), 1997-2003

(Dependent variable is the log of operating income plus interest expense)

Equation	Efficiency coefficient	Rbar2	Degrees of freedom
All banks
No exogenous variable	-18.29	***	0.99	417
L/A, D/L, A/E ratios added	-18.02	***	0.99	408
L/A, D/L, A/E, C/A ratios added	-12.39	***	0.99	407
Large banks
No exogenous variable	-15.13	***	0.99	176
L/A, D/L, A/E ratios added	-16.36	***	0.99	168
L/A, D/L, A/E, C/A ratios added	-7.70	***	0.99	167
Small banks
No exogenous variable	-11.42	***	0.98	217
L/A, D/L, A/E ratios added	-11.19	***	0.98	213
L/A, D/L, A/E, C/A ratios added	-6.43	*	0.98	212

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.1/ Confidence region is two standard errors wide on each side of point estimate.

^***denotes that the coefficient is significantly different from zero at the 1 percent level.

** denotes that the coefficient is significantly different from zero at the 5 percent level.

^*denotes that the coefficient is significantly different from zero at the 10 percent level.

View Table

Table VI.6.

Measures of Revenue Efficiency of EU Banks (Relative to US banks), 1997-2003

(Dependent variable is the log of operating income plus interest expense)

Equation	Efficiency coefficient	Rbar2	Degrees of freedom
All banks
No exogenous variable	-18.29	***	0.99	417
L/A, D/L, A/E ratios added	-18.02	***	0.99	408
L/A, D/L, A/E, C/A ratios added	-12.39	***	0.99	407
Large banks
No exogenous variable	-15.13	***	0.99	176
L/A, D/L, A/E ratios added	-16.36	***	0.99	168
L/A, D/L, A/E, C/A ratios added	-7.70	***	0.99	167
Small banks
No exogenous variable	-11.42	***	0.98	217
L/A, D/L, A/E ratios added	-11.19	***	0.98	213
L/A, D/L, A/E, C/A ratios added	-6.43	*	0.98	212

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.1/ Confidence region is two standard errors wide on each side of point estimate.

^***denotes that the coefficient is significantly different from zero at the 1 percent level.

** denotes that the coefficient is significantly different from zero at the 5 percent level.

^*denotes that the coefficient is significantly different from zero at the 10 percent level.

197. Overall, the evidence suggests that differences in business models explain a substantial part of the revenue gap of EU relative to US banks. The remaining gap could reflect a lower X-efficiency. Alternatively, the gap may be due to other factors, not considered by the explanatory variables, for example, greater competition in Europe, missing capital markets that hinder more effective intermediation, or differences in risk that are not captured by differences in equity ratios. Section E explores the role of competition in more depth.

E. Competition

Gauging Competition Among Banks

201. The US banking market offers a natural benchmark for comparing competition among EU banks, given its size and level of development. Much of the literature has focused on measures of concentration to determine the amount of competition in a banking market. By one such measure, the per capita number of banks, there appears to be more competition in the US market than in the EU 15. Furthermore, to preserve competition, no bank in the United States is to have a share of the market for deposits that exceeds 30 percent in a single state or 10 percent nationwide. The largest banks are far away from a 10 percent limit in the US but not in some smaller EU states. However, such measures of competition can be misleading for various reasons. For example, contestability might be more important than concentration. Alternatively, many banks might be operating as a group, not competing in each other’s markets, as is the case, for example, for cooperative and savings banks in some EU countries.

202. The relationship between a bank’s costs and revenues provides better information on competition than standard indicators of concentration. First, it does not require direct data of prices and comparable services, which is particularly tricky in the financial services industry. Second, there is no need to specify a geographic market. The Panzar and Rosse (1987) H-statistic—which captures this relation—is given by:

$H=\underset{i=1}{\overset{I}{\mathrm{\Sigma }}}\frac{\partial R}{\partial w_i}\frac{w_i}{R},$ where R = R(d, c, w), denotes a bank’s revenue as function of a vector of input prices w as well as exogenous variables that shift demand d or cost c. A number of standard assumptions need to be satisfied for the H-statistic to be useful, including (i) profit maximization; (ii) homothetic production functions; (iii) exogenous factor prices; (iv) an elasticity of demand that rises with the number of rivals in the market; and (v) a market that is in long-run equilibrium. Notice that conditions (ii) and (v) can potentially cause problems, notwithstanding the widespread assumption in the literature that they are satisfied. Since the analysis here focuses on the top 100 banks in each area, homotheticity should not be a major issue as these banks are fairly large and returns to scale are seen as running out at smaller levels.¹⁶² Long-run equilibrium might be a different matter, however, given the rapid pace of change in the financial services industry. Notice that under conditions (i) to (v):

• H≤0 for a monopoly market. Intuitively, any increase in cost prompts the monopolist to cut back output, which leads to a loss in revenue—the relation between cost and revenue is negative.¹⁶³

• 0<H<1 for a market characterized by monopolistic competition. An increase in a bank’s costs prompts an increase in prices but revenues do not rise one for one, as the bank’s demand curve slopes downward. Notice that a larger H-statistic implies a more elastic demand curve and thus less market power (Vesala, 1995).

• H=1 for perfect competition. If the market is perfectly competitive then there must be free entry and exit, which sets the price equal to minimum average cost; thus, any increase in cost must be matched one-for-one by revenue.

203. Implementing the Panzar-Rosse method also requires that an assumption be made about banks’ activities. As in Section D, the “intermediation approach” is followed here. Accordingly, the following regression is run:

$re{v}_{it}=\alpha +{\beta }_{1}per{s}_{-}{\exp }_{it}+{\beta }_2{\mathrm{int}}_{-}{\exp }_{it}+{\beta }_{3}ot{h}_{-}{\exp }_{it}+{\beta }_{4}asset{s}_{it}+\underset{j=1}{\overset{2}{\mathrm{\Sigma }}}{\partial }_j{D}_j+\underset{j=1}{\overset{J}{\mathrm{\Sigma }}}{\gamma }_j{Z}_{jit}+{ϵ}_{it},$

where the subscripts i and t denote bank i at time t; rev denotes operating revenue, excluding interest expenses; pers_exp personnel expenditure divided by employment; int_exp interest expenditure divided by liabilities; oth_exp is other expenditure divided by assets. These variables, including total assets, are in natural logarithms. The H-statistic is given by: H = β₁+ β₂+ β₃. The exogenous variables Z are the same as those in Section D. The time dummies D proxy for changes in the macroeconomic environment.

204. The key finding is that the small EU banks behave less competitively both relative to large EU banks and small US banks. The estimate for the H-statistic for the full sample of EU banks is about 0.5 while that for US banks is about 0.7 (Table VI.7). The confidence intervals permit the rejection of the hypotheses of pure monopoly or perfect competition in both cases, suggesting that monopolistic competition prevails. The results are in line with the findings of Brunner and others (2004), who used a much larger sample of banks and a similar estimation methodology for single EU countries, as well as with other findings in the literature for the United States.¹⁶⁴ Standard test statistics point to similar competition among large EU and US banks (Table VI.8). However, the smaller EU banks appear to behave significantly more monopolistically than their US counterparts.¹⁶⁵ Furthermore, at the 10 percent significance level large banks exhibit a higher H-statistic than their smaller counterparts in the EU (Table VI.9). The reverse appears to be the case in the US banking industry.

Table VI.7.

Measures of Competition for EU Banks and US Banks, 1997-2003

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

Table VI.7.

Measures of Competition for EU Banks and US Banks, 1997-2003

Equation	H-Statistic	Confidence region 1/
		lower bound	upper bound
All banks
		European Union
No exogenous variable	0.50	0.43	0.58
L/A, D/L, A/E ratios added	0.53	0.46	0.60
L/A, D/L, A/E, C/A ratios added	0.54	0.48	0.61
		United States
No exogenous variable	0.73	0.64	0.82
L/A, D/L, A/E ratios added	0.71	0.60	0.82
L/A, D/L, A/E, C/A ratios added	0.68	0.58	0.79
Specification 4	0.48	0.36	0.60
Large banks
		European Union
No exogenous variable	0.64	0.52	0.76
L/A, D/L, A/E ratios added	0.65	0.52	0.77
L/A, D/L, A/E, C/A ratios added	0.68	0.57	0.80
		United States
No exogenous variable	0.71	0.58	0.85
L/A, D/L, A/E ratios added	0.60	0.43	0.77
L/A, D/L, A/E, C/A ratios added	0.50	0.32	0.68
Small banks
		European Union
No exogenous variable	0.54	0.42	0.67
L/A, D/L, A/E ratios added	0.53	0.41	0.65
L/A, D/L, A/E, C/A ratios added	0.63	0.51	0.74
		United States
No exogenous variable	0.84	0.69	0.98
L/A, D/L, A/E ratios added	0.83	0.68	0.99
L/A, D/L, A/E, C/A ratios added	0.83	0.68	0.98

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

View Table

Table VI.7.

Measures of Competition for EU Banks and US Banks, 1997-2003

Equation	H-Statistic	Confidence region 1/
		lower bound	upper bound
All banks
		European Union
No exogenous variable	0.50	0.43	0.58
L/A, D/L, A/E ratios added	0.53	0.46	0.60
L/A, D/L, A/E, C/A ratios added	0.54	0.48	0.61
		United States
No exogenous variable	0.73	0.64	0.82
L/A, D/L, A/E ratios added	0.71	0.60	0.82
L/A, D/L, A/E, C/A ratios added	0.68	0.58	0.79
Specification 4	0.48	0.36	0.60
Large banks
		European Union
No exogenous variable	0.64	0.52	0.76
L/A, D/L, A/E ratios added	0.65	0.52	0.77
L/A, D/L, A/E, C/A ratios added	0.68	0.57	0.80
		United States
No exogenous variable	0.71	0.58	0.85
L/A, D/L, A/E ratios added	0.60	0.43	0.77
L/A, D/L, A/E, C/A ratios added	0.50	0.32	0.68
Small banks
		European Union
No exogenous variable	0.54	0.42	0.67
L/A, D/L, A/E ratios added	0.53	0.41	0.65
L/A, D/L, A/E, C/A ratios added	0.63	0.51	0.74
		United States
No exogenous variable	0.84	0.69	0.98
L/A, D/L, A/E ratios added	0.83	0.68	0.99
L/A, D/L, A/E, C/A ratios added	0.83	0.68	0.98

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

Table VI.8.

Measures of Competition for EU Banks (Relative to US Banks), 1997-2003

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

Table VI.8.

Measures of Competition for EU Banks (Relative to US Banks), 1997-2003

Equation	EU banks' H-Statistic		Confidence region 1/
	(Relative to US banks)		lower bound	upper bound
All banks
No exogenous variable	-0.23	***	-0.34	-0.13
L/A, D/L, A/E ratios added	-0.18	***	-0.31	-0.05
L/A, D/L, A/E, C/A ratios added	-0.14	**	-0.26	-0.02
Large banks
No exogenous variable	-0.08		-0.27	0.10
L/A, D/L, A/E ratios added	0.04		-0.17	0.25
L/A, D/L, A/E, C/A ratios added	0.18	*	-0.03	0.39
Small banks
No exogenous variable	-0.34	***	-0.53	-0.16
L/A, D/L, A/E ratios added	-0.34	***	-0.53	-0.15
L/A, D/L, A/E, C/A ratios added	-0.23	***	-0.42	-0.04

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

View Table

Table VI.8.

Measures of Competition for EU Banks (Relative to US Banks), 1997-2003

Equation	EU banks' H-Statistic		Confidence region 1/
	(Relative to US banks)		lower bound	upper bound
All banks
No exogenous variable	-0.23	***	-0.34	-0.13
L/A, D/L, A/E ratios added	-0.18	***	-0.31	-0.05
L/A, D/L, A/E, C/A ratios added	-0.14	**	-0.26	-0.02
Large banks
No exogenous variable	-0.08		-0.27	0.10
L/A, D/L, A/E ratios added	0.04		-0.17	0.25
L/A, D/L, A/E, C/A ratios added	0.18	*	-0.03	0.39
Small banks
No exogenous variable	-0.34	***	-0.53	-0.16
L/A, D/L, A/E ratios added	-0.34	***	-0.53	-0.15
L/A, D/L, A/E, C/A ratios added	-0.23	***	-0.42	-0.04

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

Table VI.9.

Measures of Competition for Small EU Banks (Relative to large EU Banks), 1997-2003

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

^*denotes that the coefficient is significantly negative at the 10 percent level.

Table VI.9.

Measures of Competition for Small EU Banks (Relative to large EU Banks), 1997-2003

Equation	Small banks' H-Statistic		Confidence region 1/
	(Relative to large banks)		lower bound	upper bound
No exogenous variable	-0.12	*	-0.29	0.05
L/A, D/L, A/E ratios added	-0.13	*	-0.30	0.04
L/A, D/L, A/E, C/A ratios added	-0.07		-0.23	0.09

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

^*denotes that the coefficient is significantly negative at the 10 percent level.

View Table

Table VI.9.

Measures of Competition for Small EU Banks (Relative to large EU Banks), 1997-2003

Equation	Small banks' H-Statistic		Confidence region 1/
	(Relative to large banks)		lower bound	upper bound
No exogenous variable	-0.12	*	-0.29	0.05
L/A, D/L, A/E ratios added	-0.13	*	-0.30	0.04
L/A, D/L, A/E, C/A ratios added	-0.07		-0.23	0.09

Sources: Fitch IBCA database; and IMF staff calculations. L stands for loans; A for assets; D for deposits; and C for capital.

^1/Confidence region is two standard errors wide on each side of point estimate.

^*denotes that the coefficient is significantly negative at the 10 percent level.

205. The results suggest that the smaller, less-internationally-oriented banks appear to operate under more sheltered conditions in the EU. The literature provides further evidence on this. Bikker and Haaf (2002), for example, using data on 23 countries find that competition is weaker among small banks—operating in local markets—than among large banks—operating predominantly in international markets, while medium-sized banks take an intermediate position.¹⁶⁶ They also show evidence for the conventional wisdom that competition and concentration are inversely related, although this finding does not receive unambiguous support in the literature, which stresses the importance of contestability.¹⁶⁷ Furthermore, Guevara, Maudos, and Perez (2005) show that the problems with competition in European banking markets are more pronounced in the retail sector, indicating that national entry barriers continue to exist.

206. Losses related to market power have been found to be large in the EU banking system. Guevara and Maudos (2004, 2005), for example, estimate the welfare losses in the European banking system related to excessive market power at the equivalent of between 1½–2½ percent of GDP. Evidence in the literature suggests that subjecting banks to more competition may have beneficial effects for household and firms.¹⁶⁸ In that sense, allowing cross-border mergers and acquisitions as well as foreign entry would produce higher growth and welfare.

F. Efficiency, Competition, and Financial Market Structure

207. More competition as a reason for lower bank revenues in Europe is not an explanation that is consistent with the empirical evidence presented above. Perfect competition among banks does not appear to be the rule either in the European Union or the United States. However, judging by per capita employment in banking, the number of banks, the competition from other sources of funds, and the results of more sophisticated techniques to gauge competition (the H-statistics), EU banks—particularly the relatively smaller, more nationally-oriented ones—appear to operate in a more sheltered environment.

208. Various other factors could explain the lower revenue effectiveness of EU banks. This section briefly explores potential explanations, including: (i) less financial innovation because of less competition; (ii) less market exit; (iii) more stringent laws and regulations governing the supply of financial services; (iv) less scope for reaching a broad base of customers; (v) more public sector intervention; and (vi) missing, complementary capital markets that would allow banks to specialize further, fostering the adoption business models that are more suited to their comparative advantages. All these explanations are consistent with the evidence on the revenue gap of EU banks presented in Section D and it is difficult to assess their relative importance.

209. Less financial innovation because of less competition: Regarding innovation, with less competition the pressure among banks to come up with new financial services is likely to be lower. Relatedly, there might be less of an incentive for markets to come up with new sources of funding. Interestingly, Altunbas and Marquéz Ibáñez (2004) find that over 1992–2001, bank mergers in the EU have led to improved returns on capital, particularly in cross-border cases.

210. Less bank turnover, particularly less market exit: The issue is not necessarily excess capacity, as argued by many, but an ineffective use of existing capacity. Controlled market exit goes hand in hand with more competition in fostering efficient bank business.¹⁶⁹

211. More stringent laws and regulations: In many countries, legal or regulatory obstacles hinder the supply of a broad range of mortgage products. Similarly, usury laws might inhibit the emergence of a broader market for consumer credit. And tax laws, for example, might discourage securitization, as was the case until recently in Germany.

212. Less scope for reaching a broad customer base: Differences in the legal and regulatory environment across countries hamper the provision of financial services to firms and households across national borders. This may reduce the payoff to innovation and thus revenues.

213. More public sector intervention in EU banking: Intervention can be explicit—through ownership of credit institutions, or implicit—by influencing the decision making of major banks, notably those that were formerly publicly-owned. Explicit intervention is still fairly widespread in Europe, although much less so among the sample of banks considered here. Furthermore, cooperative banks—which, although not state-owned, do not necessarily have profit maximization and innovation as their primary objective—are more widespread in Europe.

214. Missing or less-developed markets and thus a less efficient division of tasks between banks and markets: EU banks absorb in their balance sheets financing activities that US banks typically channel through financial markets and instruments. These activities likely require less specialized banking knowledge and are probably less risky, generating lower income streams and requiring less capital. The results in Section D show that upon holding constant for the lower capital held by EU banks, the revenue gap relative to their US counterparts shrinks considerably. Nonetheless, a gap remains. Be that as it may, EU banks likely have a larger proportion of assets in their balance sheets that, in the future, could be sold off in markets, including, for example, large corporate loans or mortgage related lending to households, both of which are largely off US banks’ balance sheets. Thus, what is captured here as a lower revenue effectiveness might merely be a reflection of missing, complementary capital markets in Europe, where securitization, for example, has gained a strong foothold in a few countries only.¹⁷⁰

215. The interplay between government intervention, market forces, and the regulation of financial activity might be at the root of the different allocation of tasks between banks and markets in Europe and the United States.¹⁷¹ In general, government intervention in Europe was relatively less market- and more bank-friendly than in the United States. Many European authorities entered credit markets directly, via ownership of a large number of banks and thus their intervention can be considered bank friendly. In the United States, the public sector played a crucial role in developing the market for securitization, notably with the introduction of mortgage-backed securities by Ginnie Mae in the 1960s.¹⁷² Securitization along-side the integration of state banking markets has played an important role in integrating regional housing markets in the United States (^{Box VI.1}).

G. Conclusions and Policy Implications

216. A long history of fragmentation means that much sand remains in the wheels of Europe’s financial system, notwithstanding significant progress recently. Aside from the degree of political union, the absence of a single currency in Europe until recently is perhaps the crucial factor. As a result, money and capital markets and their regulatory and supervisory infrastructure were, for a long time, highly fragmented in Europe. Fragmentation remains an issue today, including both actual and perceived obstacles to crossborder activities.¹⁷³ This fragmentation comes at the price of a less efficient and resilient financial sector. On the latter, available evidence for the United States might offer some useful lessons: among the causes of the banking crises of the 1980s feature laws that inhibited competition, geographic diversification of risks, and consolidation of units.¹⁷⁴

217. The result is a financial system in Europe that presently appears to offer less scope for banks and markets to leverage their comparative advantages. Capital markets are smaller in Europe. In relative terms, banking sector employment is similar in Europe and the United States. However, EU banks are doing business differently. Their balance sheets are larger, less risky, and they hold less equity. The key reason is that they engage less in placing risks in financial markets than their US counterparts. In many countries, the types of markets that are necessary for banks to pursue such activities, for example, securitization, are only in their infancy.

Box VI.1.

Securitization and the Integration of Local Housing Markets

Both the integration of banking markets and securitization played a role in reducing the divergence of house price increases across the nine OFHEO (Office of Federal Housing Enterprise Oversight) regions in the United States.

The integration of state banking markets is evidenced by the increase in the weighted average of interstate asset ratios (i.e., the percent of bank assets held by out-of-state bank holding companies). Morgan, Rime, and Strahan (2004) show that interstate banking has made state business cycles smaller.

Concomitantly, securitization of mortgage loans rose rapidly and the share of deposits that fund mortgages fell significantly—from over 70 percent to less than 40 percent recently—as shown by Schnure (2005). He establishes a significant negative relation between the share of securitized mortgages on the one hand and the cross-sectional standard deviation of increases in the OFHEO house price indices (for nine regions) on the other hand.

Interestingly, notwithstanding much lower interregional migration, house prices appear to diverge more across the EU countries than across the nine OFHEO regions in the United States. Nonetheless, the BIS data show falling divergences over time.

• Download Figure
• Download figure as PowerPoint slide

View Full Size

• Download Figure
• Download figure as PowerPoint slide

• Download Figure
• Download figure as PowerPoint slide

• Download Figure
• Download figure as PowerPoint slide

View Full Size

• Download Figure
• Download figure as PowerPoint slide

• Download Figure
• Download figure as PowerPoint slide

• Download Figure
• Download figure as PowerPoint slide

View Full Size

• Download Figure
• Download figure as PowerPoint slide

• Download Figure
• Download figure as PowerPoint slide

218. A balanced integration of EU banking and capital markets would be desirable. This will require coordination among market players and some policy intervention by public authorities. The evidence here would support action on two fronts:

• Governments may have to play a role in fostering the development of new financial markets. The Commission’s Green Paper (2005) recognizes that Europe’s capital market is underdeveloped. The current setting, with significant legal and regulatory differences across EU countries, fragmented clearing and settlement systems, and continued government intervention in banking is likely to hinder the development of markets. Banks are better equipped to operate in a less homogenous regulatory environment. But an integration that is skewed toward banks rather than markets may well be less beneficial for firms and households.

• Governments should foster crossborder banking to boost competition and incite banks to pass along effectiveness gains to their customers. The Green Paper rightly emphasizes that competition policy is an important complement to financial integration measures. Europe’s internationally active banks already appear to engage in more competition than their smaller counterparts. Thus allowing such banks to contest new markets would foster more efficient and innovative financial intermediation. The flipside of fostering crossborder banking is to promote crossborder shopping for financial services by firms and households. The Green Paper’s objectives in this domain should be welcomed. More specifically, broadening the range of mortgage products available to households—an issue to be covered in a future Green Paper—would help in integrating this important market. Ideally, it should be supported with an integrated market for securitization.

219. A wider array of markets and more competition among banks offer a number of benefits for the economy but developments in this direction need to be monitored. Key among the benefits are (i) a more efficient use of bank capital; (ii) better risk management; (iii) a greater resiliency of the financial system to sudden increases in the demand for liquidity; and (iv) an improved pass-through of monetary policy to the real economy. Ultimately, the efficiency gains accrue to real economy and to the consumer in the form of a higher returns on savings, which is crucial in the context of an aging society. However, the transition to a more complete and complex financial system in an environment of rising competition among financial intermediaries will require special vigilance by supervisors.