You are on page 1of 65

Conference on Financial Modernisation and Economic Growth

in Europe
Berlin, 28-29 September 2006

Philipp Hartmann (European Central Bank),


Annalisa Ferrando (European Central Bank),
Friedrich Fritzer (Oesterreichische Nationalbank),
Florian Heider (European Central Bank),
Bernadette Lauro (European Central Bank) and
Marco lo Duca (European Central Bank)

„The Performance of the European Financial


System“

http://www.bundesbank.de
The Performance of the European Financial System*

Philipp Hartmann§, Annalisa Ferrando†, Friedrich Fritzer‡, Florian Heider†, Bernadette


Lauro† and Marco Lo Duca†

§
European Central Bank and CEPR

† ‡
European Central Bank Österreichische Nationalbank

Revised and extended draft, August 2006

Abstract: The present article studies the quality with which financial systems perform their main
functions. The better financial systems perform these functions, the more they contribute to
productivity and economic growth. While the expanding theoretical and empirical economic
literature that emphasises this relationship has focused mostly on developing economies, this
article concentrates on industrialised economies with relatively developed financial systems. A
comprehensive framework for assessing financial system performance is established, which leads
to the identification of eight dimensions along which the performance of a financial system can be
characterised. For each of these dimensions the article presents a series of indicators (41 in total)
that can be used to gauge the quality with which a financial system performs its main functions.
In so doing, financial systems in euro area countries are compared to other major industrialised
countries, both inside and outside of Europe. The framework and results are relevant for the
ongoing debate about structural reforms in European Union countries, in particular how financial
sector reforms can contribute to policies stimulating growth and international competitiveness.

Key words: Financial systems, financial development, financial markets, financial intermediaries,
economic indicators
JEL classification: G1, G2, G3, K2, O16
* We are very grateful to Asli Demirgüc-Kunt and Franklin Allen for deep discussions about
indicators of financial system performance and to Ross Levine for further discussions and help to
trace down data. Fritzer conducted part of the project while visiting the ECB. Any views
expressed are only the ones of the authors and do not necessarily represent official views of the
ECB, the Austrian Nationalbank or the Eurosystem.

1
1. Introduction

The financial system is an important part of the economy. It ensures that savings are channelled to
the most productive investment opportunities and it allows for inter-temporal and cross-sectional
risk sharing (see for example Allen & Gale (2000)). In so doing, it helps households to smooth
consumption and firms to grasp investment opportunities when they arise. Hence, a well-
functioning financial system enhances welfare and is conducive to economic growth. A growing
body of literature supports these predictions with empirical evidence (see, e.g., Demirgüç-Kunt
and Levine (2004) or Levine (2005) for a recent survey of the literature). In particular, this
literature argues that improvements in the functioning of financial systems are an inherent part of
the development strategies of less developed, transition and emerging market countries. Much
less clear from this literature is, however, the role of financial sector reforms in industrial
countries with already relatively developed financial systems.

The functioning of financial systems has received special attention in European public policy in
recent years. A well-functioning financial system permits an economy to fully exploit its growth
potential, as it ensures that the best real investment opportunities receive the necessary funding.
Therefore, the European Union has made structural financial sector reforms one priority under the
Lisbon Agenda. In particular, in the EU Commission’s Financial Services Action Plan (1999-
2004), which is currently implemented in member countries, greater European financial
integration to complete the single market for financial services has been an important objective.
The Commission has also published a White Paper, explaining its financial sector policies for the
period 2005 to 2010. From the perspective of central banks, financial sectors also play crucial
roles in the implementation and transmission of monetary policy. This is a primary reason why
the ECB has a special interest in the functioning of the euro area financial system (see for
example Gaspar et al (2002)).1 Finally, the functioning of financial systems is relevant for
financial stability, which is of importance for both supervisory authorities and central banks.

In this context, this article presents an analysis of the performance of financial systems that is
novel in various respects. In particular, the approach adopted is broad in that it encompasses the

1
The interest of the ECB in financial sector issues is also reflected in the Monthly Bulletin articles “Recent
developments in financial structures of the euro area” (October 2003) and “Assessing the performance of
financial systems” (October 2005), publication of the bi-annual ECB Financial Stability Review since
December 2004 as well as the first report on “Indicators of financial integration in the euro area”
(September 2005).

2
typical issues of financial structure and integration but goes much beyond those. The reason is
that some evidence seems to suggest that other dimensions of the performance of financial
systems, often summarised under the terms financial development or modernisation, may be even
more relevant for economic growth (see for example Levine and Zervos (1998), Edison et al
(2002) and Giannetti et al. (2002)). Our comprehensive approach starts from the well established
functions of a financial system and covers all of its dimensions that are likely to affect growth.
For each dimension a group of economic indicators can be derived that describe the quality with
which a given financial system fulfils its functions. This article presents a selection of indicators,
some of which are new while others are indicators that have been used in the finance and growth
literature before (see e.g. those described in the surveys by Demirgüç-Kunt and Levine (2001) or
Levine (2005)) and that we update where possible using previously unavailable data.2 In contrast
to most of the finance and growth literature, the focus here is on industrialised countries, covering
euro area countries and comparing them to major countries inside and outside of Europe. By
looking at a large number of indicators, a more reliable assessment of the overall performance of
a financial system can be made.

The next section introduces the main functions of a financial system and reviews the economic
theory about how market imperfections and frictions can reduce its contribution to growth. In
order to see in more detail how market imperfections lead to an inefficient use of scarce
resources, we first present a simple reduced form model of corporate investment. We then
identify groups of characteristics/dimensions that lend themselves to an assessment of how well a
financial system performs its functions. In particular, we select eight dimensions along which we
structure our subsequent analysis, including financial innovation and market completeness,
transparency and information, corporate governance, legal system, regulation and supervision,
competition and openness. The third section presents the results for the indicators and discusses
their interpretation. Wherever possible, we display an indicator for each of the 12 euro area
countries and the euro area aggregate, for other European countries (Sweden, Switzerland and the
United Kingdom) and for major non-European countries (Japan and the United States). To
identify trends, we track indicators over time, subject to data availability. The comparison with
non-euro area countries allows identifying those trends that are specific to the euro area and those
that are not. The last section draws some conclusions.

2
We use data that is homogenous across countries so that one can make cross-country comparisons. Such
international data may not always coincide with particular national data that do not allow making these
comparisons.

3
2. Theory and dimensions of financial system performance

A financial system influences the allocation of scarce resources across space and time. In order to
exploit the growth potential of an economy, resources need to be reallocated towards the best
investment opportunities. When frictions hinder the flow of resources to the best possible uses,
economic growth suffers. A financial system addresses these market imperfections since, it

• produces information about investments and possibilities to allocate capital,

• monitors investments and ensures that investors and savers are paid back according to the
contracts they hold,

• allows the trading, diversification and management of risk,

• mobilises and pools savings,

• eases the exchange of goods and services. 3

In order to see in more detail how market imperfections lead to an inefficient use of scarce
resources, we first present a simple reduced form model of corporate investment.4 The model
illustrates our discussion of how a financial system functions to overcome these frictions and thus
affects growth.

3
We follow the classification of Levine (2005) here.
4
While the model illustrates a number of particularly important elements in the working of financial
systems, it does not capture all their functions in an exhaustive way. For example, it does not formalise
household saving decisions. There is a literature that suggests that physical capital accumulation per se does
not account for all that much of long-run economic growth (Jorgenson, 1995 and 2005). The model also
does not explicitly treat the allocation of risk. The literature argues that the effect of risk on savings is
ambiguous. This is due to the interplay of the well known income and substitution effects (see, e.g., Levhari
and Srinivasan, 1969). Instead, the model treats saving and risk in reduced form via the supply curve of
capital. See also Levine and Zervos (1998) who find that stock market volatility is not robustly linked with
growth and that private savings rates are not closely associated with indicators of financial development.

4
2.1 The model5

The simple model first illustrates the benchmark case that occurs when capital markets are
perfect, and there is no wedge between the internal and external cost of capital. In a second step
we discuss deviations from the perfect market benchmark leading to inefficient investment
decisions.

There are two dates. At time zero, the firm is endowed with a stock of internal funds F. It then
invests an amount I that yields a future gross return of f(I) at date one. The production function
f(.) is an increasing, concave function. After normalising the risk-free interest rate to one, the
profit from investing is then f(I)-I. The first best level of investment I* maximises profits. It
occurs when the marginal productivity equals the marginal cost of capital, which is equal to one
since this is the rate that funds can earn if they are not invested:
f '(I ) = 1. (1)

The firm can raise an amount X of external capital at date 0, but it must repay an amount p(X)X to
outside investors at date one, where p(X) is price of external financing of size X. The firm’s
budget constraint is

I≤X +F (2)

The firm’s investment problem then is


max f ( I ) − I + X − p( X ) X (3)
I,X

subject to the budget constraint (2).

Perfect market benchmark

Economic theory states that a perfect market is characterised by a frictionless flow of capital
ensuring that all valuable investment opportunities are exploited optimally. Even though some
agents transfer their capital, and thus give up control, to others who may have different
information or different interests when investing, it is possible to specify at no cost contracts that
cover all possible future contingencies. Similarly, households can achieve optimal consumption
smoothing and risk sharing over their lifecycle. In a perfect market where capital flows without

5
The model is based on Kaplan and Zingales (1997) and Stein (2001).

5
frictions, the organisation of economic activity, i.e. firms, institutions and the location of
economic agents, is irrelevant (see for example Coase (1937) or Modigliani Miller (1958)). It
does not matter whether savers and investors are the same person or not. A perfect market
achieves the same allocation of capital and level of investment as if each owner of an investment
opportunity, e.g. a firm or a household, was already endowed with sufficient resources to invest
up to the optimal point where the marginal benefit equals the opportunity cost so that there is no
need to contact capital markets.

In the model, the perfect markets benchmark is achieved when p(X)=1, i.e. when the supply of
capital is perfectly elastic and the cost of capital is equal to the marginal cost of internal funds.
Then the investment problem (3) yields the first best level of investment in (1). Frictionless
external financing closes any gap between the first best level of investment I* and the amount of
internal funds F.

Capital market imperfections

It is clear that in reality firms and households need to contract in financial markets that are not
perfect and frictionless. First, economic agents neither share nor have access to the same
information. Investors, for example, provide the investment capital but delegate the investment
decision to a manager, since he often has better information about the use of capital. When an
investor no longer has control over his funds, he demands a premium that increases the cost of
capital, since he needs to be compensated for not knowing exactly how his funds are being used
by the manager. The increase in the cost of capital makes investing more expensive and leads to
underinvestment relative to the perfect market benchmark (see for example Stiglitz and Weiss
(1981) and Myers and Majluf (1984)).

In the model, we can illustrate this capital market imperfection by setting the price of external
capital larger than the cost of internal financing, p>1. When the budget constraint (2) is binding,
the first order condition for the investment problem (3) becomes:6
f ' ( I ) = p + p' X (4)

The firm invests until the marginal benefit equals the marginal cost of external financing. If the
cost of external financing p(X) increases, the firm invests less:

6
If the investment is of such a small scale that it can be financed internally and the budget constraint does
not bind, then imperfections in external capital markets clearly do not matter.

6
dI 1
= <0
dp f ' ' (5)

The price sensitivity of investment is larger when the production function f is more curved, i.e.
when changes in investment change the profitability of the firm a lot. The underinvestment
problem created by asymmetric information is therefore more severe for small, young or risky
firms.

Bank financing and specialised venture capital financing are responses of a financial system to
this kind of information friction, especially for small and medium enterprises. In both cases,
uninformed investors hand over their resources to more knowledgeable intermediaries, who have
learnt from financing similar projects. In addition, they pool the resources of many investors and
can therefore reach a sufficient scale to cover the fixed costs of acquiring information. These
fixed costs originate, for example, from setting up a credit screening mechanism and from the
“public goods” problem, i.e. the difficulty of preventing others from free riding on one’s costly
information. Stock markets support the acquisition and dissemination of information about firms
mainly through the price mechanism. Trading in a large and liquid market means that prices
reflect the information of many traders who have an incentive to acquire valuable information.

A second type of capital market imperfection is due the to separation of ownership and control in
modern corporations. A financial investor – who owns the funds – and his manager – who de
facto controls them – may not share the same investment objectives. While investors are usually
interested in value maximisation, as in the perfect market benchmark, managers may be driven by
career concerns or perks, or, in extreme cases, they may even extract resources for themselves
(see Jensen and Meckling (1976)). In the same vein, the interests of a household may not be
identical to those of the bank it borrows from, since a bank is only interested in the repayment of
its loans, or the interests of a mutual fund manager may be different from the ones of the
households whose money he invests. When investment decisions are no longer governed by value
maximisation, they will be distorted relative to a perfect market and growth will be hampered
since scarce capital is not used efficiently.

We can illustrate the effect of this type of market imperfection on firms’ investment decisions in
our simple model by rewriting (3) as
max γf ( I ) − I + X − p( X ) X (6)
I,X

7
where γ is a parameter that describes the divergence between management’s and owners’
objectives. When γ>1 (γ<1) there is going to be over (under)-investment. The classic argument
for γ<1 is the “moral hazard” problem in which managers exert too little costly effort, either
because of insufficient incentives or because they are too risk averse. The standard argument for
γ>1 is that managers may enjoy “private benefits of control” that are proportional to the level of
investment, such as the use of a corporate jet for example.

Several arrangements in a financial system address the control problem. Banks for example serve
as “delegated monitors”, i.e. a bank acts on behalf of many depositors, thus avoiding a wasteful
duplication of monitoring expenses (see Diamond (1984)). Stock markets allow investors to exert
pressure by selling their shares. The buying or selling of shares influences a company’s stock
price with possible consequences for management, e.g. its dismissal after a poor stock price
performance. More directly, stock markets exercise control through voting at shareholder
meetings or, in extreme cases, through takeovers.

A third type of friction in financial markets is that capital is dispersed among many different
people who have different time and risk preferences (see for example Allen and Gale (1997)).
The perfect market benchmark assumes that capital is fully liquid, meaning that financial assets
can be traded and converted into real assets without frictions. In reality, physical assets used in
production and human capital are highly illiquid, e.g. it is neither easy to buy or sell production
plants, nor to borrow against future expected income.

To finance production, capital must be committed in many cases for a long period of time, but a
saver does not typically like relinquishing control over his assets for long periods. A saver’s
investment horizon is often shorter than the investment horizon of industrial production.
Households are subject to liquidity risks, i.e. they may need quick access to their capital in order
to cover unforeseen contingencies. Banks and stock markets mitigate the adverse effects of such
liquidity risks. A bank transforms short-term liquid deposits into long-term illiquid loans (see
Diamond and Dybvig (1983)). A bank therefore makes it possible for households to react to
liquidity shocks and to withdraw deposits without interrupting industrial production. Stock
markets similarly reduce liquidity risks by allowing stock holders to trade their shares, while
firms still have access to long-term capital.

8
Fourth, in the perfect market benchmark, the location of capital and economic activity does not
matter since financial contracts can be written to cover all possible future contingencies. But in
reality economic capital is dispersed across many investors. Without access to multiple investors,
many production processes would be constrained to sub-optimal scales. A key function of a
financial system is therefore the pooling and the mobilisation of scarce and dispersed capital.
Having standardised financial contracts, such as bonds or shares, lowers the transaction costs of
trading in public markets. Without such standard contracts, firms would have to enter into a large
number of bilateral agreements, specifying a large number of contingencies, instead of being able
to tap into a large pool of readily available capital. A pooling of resources also occurs through
financial intermediaries, where a large number of depositors entrust their funds to a “middleman”
who is then able to invest on a large scale on their behalf.

Finally, and closely related to the previous point about the location of economic activity, a
financial system facilitates the exchange of goods and services. In order to exploit the full growth
potential of an economy, specialised investments need to be made and households need to be able
to finance the consumption of goods. Greater specialisation allows higher returns to be earned,
because it makes better use of the information and skills that are specific to a production process.
But greater specialisation also requires more coordination and transactions than an autarkic
environment. At the same time, the financing of consumption over time also requires financial
arrangements. Households for example need to be able to save or to borrow against their future
stream of income.

In the perfect market benchmark, goods and services flow without frictions across production
processes and to consumers. Information asymmetries and transaction costs, however, cause
frictions to build up in this exchange. A financial system overcomes these barriers by providing
suitable specialised instruments such as derivatives, which can fix prices in advance (e.g. forward
contracts). Credit cards, consumer credit and mortgage refinancing are channels through which a
financial system facilitates the consumption of goods and services by the household sector.

These last three frictions, risk diversification, mobilising savings and easier exchange of goods,
are somewhat outside the scope of our simple model. Nevertheless, the shape of the supply curve
of external capital p(X) does reflect these frictions to some extent. For example, less diversified
households demand a higher price for parting with their savings for long periods of time. The
same holds when firms have to pay large transaction and contracting costs to attract scarce capital

9
that is scattered among many different investors. Larger frictions result in a more inelastic supply
of capital, i.e. p’(X) increases, which in turn pushes up the marginal cost of capital in (4).

To sum up, a financial system performs several functions to ease the flow of resources from
owners of capital, e.g. households which possess savings, to ensure efficient investment, and
hence growth, in an economy when there are a number of market imperfections.

2.2 Dimensions of financial system performance

Having reviewed the main elements of economic theory explaining how a financial system helps
to overcome the frictions that can hinder the efficient flow of savings into investment, one can go
ahead and identify groups of characteristics/dimensions that lend themselves to an assessment of
a financial system’s performance of its functions. Table 1 summarizes the eight dimensions along
which we structure our subsequent analysis. For each dimension, we present a number of
indicators that measure a particular aspect of the functioning of financial systems. We draw on
the existing empirical evidence on the link between finance and growth, but present also a
number of new indicators, and we pay special attention to dimensions and indicators that are
particularly relevant for industrialised countries. Where possible, the analysis shows the evolution
of an indicator over time, distinguishing euro area countries and the euro area aggregate from
non-euro area European countries and other major countries.

10
Table 1: Dimensions of financial system performance covered by the indicators

1. Size of capital markets Financial systems with larger overall capital markets will
and financial structure provide easier financing for real investment. This relates to both
larger securities markets and to more bank credit. Systems that
rely primarily on one but not the other may be less efficient.
Also the liquidity of the different markets is relevant for this
dimension.
2. Financial innovation and Many financial innovations reduce capital market imperfections
market completeness and make markets more complete. This opens up new
possibilities to allocate capital across space, time and risk
preferences. New financial instruments and practices for
example allow firms to manage certain risks by shifting them to
investors who have a better ability to bear them.
3. Transparency and Financial systems help produce and spread information about
information investment opportunities, market conditions and the behaviour of
agents. The better they function, the lower should e.g. be
asymmetric information between firms and outsiders and the
more information should be incorporated into stock prices.
4. Corporate governance There are conflicts between insiders who control a firm and
outside investors who provide financing. Better governance will
ensure that investors receive the full return on their investment
and that there will be little deadweight costs due to opportunistic
behaviour by firm insiders, with beneficial effects on the cost of
capital.
5. Legal system A key aspect of a financial system is how well it enforces
contracts. As it allocates capital across time and space, contracts
– either explicit or implicit – are needed to connect providers and
users of funds. The legal system and how it is applied by legal
institutions determine the “distance” over which capital can be
reallocated.
6. Financial regulation, Government intervention in financial systems tends to be
supervision and stability stronger than in other economic sectors. Well designed
regulation and supervision should correct for market
imperfections and enhance stability, whereas imperfect policies
may have adverse effects on the performance of the financial
sector.
7. Competition, openness More openness of a financial system and more competition
and financial integration among banks and other financial intermediaries lower capital
market imperfections. Pressure from competition, for example,
should ensure that financial institutions operate efficiently, earn
fewer rents from market power and provide new instruments to
customers.
8. Economic freedom, Economic freedom means the absence of constraints to economic
political and socio- activities, e.g. corruption, administrative burdens or not
economic factors efficiency-related political interventions. Given the great
importance of information, contract enforcement and ease of
exchange in financial transactions, there is also a significant role
for social capital in the form of cooperativeness, ethics and trust.

11
3. Indicators of financial system performance

3.1. Size of capital markets and financial structure

Financial structure traditionally refers to the relative size of different capital markets, e.g.
securities markets. There has been an extensive debate on whether bank-based or market-based
financial systems perform better. On the one hand, banks can exploit scale economies in
acquiring information, exercise control through monitoring, and form long-run relationships with
firms that reduce information asymmetries. On the other hand, banks may exploit their dominant
position vis-à-vis a borrower, they may have a bias towards prudence, and they maximise their
own interests and do not necessarily act in the interest of firm owners. More recently, the debate
has however shifted away from pitting markets against banks. Rather, it is recognised that both
perform valuable functions in a financial system. For example, Allen and Gale (1997) argue that
banks foster inter-temporal risk sharing, whereas markets foster cross-sectional risk sharing.
Some functions may even require complementarities between bank- and market-based financing
(see for example Levine (2002)).

Therefore chart 1 displays a broad indicator of overall capital market size, measuring total
financing in the economy. It aggregates bank credit to the private sector, stock market
capitalisation and the outstanding amount of domestic debt securities issued by the private sector
as a share of GDP. Several contributions to the literature have shown that private credit or stock
market capitalisation can be important indicators of financial development (see for example King
and Levine (1993), Levine and Zervos (1998) and Rajan and Zingales (1998)). Since we are
dealing with industrial countries, however, we prefer to complement them with corporate bond
markets, which are usually only an important form of finance in already quite developed financial
systems.

Chart 1: Size of capital markets

The size of capital markets has increased since the 1990s for all countries except Japan. The
Netherlands, Greece, Portugal and to some extent Spain have seen the size of their markets more
than double over the last fourteen years. Austria, Greece and Italy have the smallest markets,
while the financial centres of Luxembourg, Switzerland and the United Kingdom have the largest

12
ones. The average size of financial markets in the euro area is roughly comparable to the size of
financial markets in Japan and it is much smaller than the United Kingdom and the United States.

A complementary measure to the overall size of capital markets is the volume of securities traded
in a market: a high volume of trading indicates a liquid market that channels large amount of
funds across investors and firms fast. High turnover is often seen as an indicator of low capital
market frictions (see Levine and Zervos (1998)). Greater liquidity usually improves price
discovery and lowers transaction costs. Chart 2 for example shows the stock market turnover, i.e.
the value of the trades of domestic shares on domestic exchanges divided by the value of listed
domestic shares.

Chart 2: Turnover ratio

Chart 2 shows that trading activity varies considerably across the euro area. France, Belgium, the
Netherlands and Portugal now trade their domestic shares via the Euronext stock exchange that
has less turnover than the Italian, German and Spanish stock market. The turnover has increased
on the most active stock exchanges, e.g. Germany and Italy, but has decreased in less liquid
markets such as Austria or Greece. No European stock market matches the turnover ratios found
on the UK and especially on the US stock market.

Taken together, charts 1 and 2 show that the size of a financial market does not necessary
correspond to its liquidity or how active trading in that market is. For example, Austria’s capital
markets are comparable in size to those in Italy, but the Austrian stock market is much less liquid
than the Italian one. Also, the capital markets in the US and the UK are of comparable size, but
trading on the US stock market is twice as active as in the UK.

The next sub-sections go beyond the traditional size measures of financial markets and consider
indicators that relate more directly to the different functions of a financial system.

3.2. Financial innovation and market completeness

Financial innovation opens up new possibilities for economic agents to allocate capital across
space, time and risk. This makes markets more complete. For example, new financial instruments

13
allow firms to manage risks by shifting them to economic agents which have a better ability to
bear them. Financial innovation allows households to refinance their mortgages and banks to
resell risks using mortgage-backed securities. Prices of new financial instruments contain
additional information. Thus, financial innovation facilitates the supply of capital and reduces
information asymmetries (see for example DeMarzo (2005)).

We propose to measure financial innovation by examining the extent to which modern financial
instruments such as asset- or mortgage backed securities, interest rate derivatives or venture
capital contracts are used. Chart 3 proposes a measure of securitisation, i.e. the transformation of
illiquid assets such as receivables or mortgages into liquid ones. The chart shows the proceeds
from the issuance of asset- and mortgage backed securities (ABS and MBS) as a fraction of
domestic GDP by country of securitised assets. For example, in 2004, assets worth a little over
7% of GDP were securitised in Germany.

Chart 3: Securitisation

The chart indicates that the level and the growth of securitisation are uneven across the euro area.
While a lot of Spanish, German, Dutch and UK assets are securitized, there is hardly any such
activity occurring in Austria, Belgium, Finland, Greece and Sweden.

Chart 4 shows the turnover of interest rate derivatives.7 More active trading of interest rate
derivatives allocates capital across time and risk because investors can lock in future interest rates
using forwards and futures, hedge using options and exchange fixed and flexible interest rate
agreements using swaps.

Chart 4: Turnover of interest rate derivatives

The most active markets for interest rate derivatives are the financial centres of Luxembourg and
the United Kingdom. The market is much less active in most other countries, although Austria,
Belgium, France and Ireland have seen strong growth of interest rate derivative trading over the
7
It would be even more informative to compare the amount of outstanding interest rate derivatives across
countries, as this would illustrate inter alia the extent of actual hedging activities. Unfortunately, such data
are not available in the format used in the chart. Moreover, ideally also more recent innovations such as
credit derivatives or structured finance products should be covered. But again no detailed country-by-
country information is available for them.

14
last decade. The market has recently shrunk in Germany, Spain and the Netherlands. On average,
interest rate derivatives are much more actively traded in the euro area than in the United States.

A caveat when interpreting chart 4 is that the location of trading may not be fully informative
about the location of counterparties or underlying assets. An investor in country A may enter into
a contractual agreement with an investor in country B, while the contract itself is traded in
country C and the underlying asset is located in yet another country. This illustrates that financial
systems can benefit from the existence of standardised financial instruments whose trading
activity transcends national borders. Chart 5 presents an example of such a standardised financial
instrument by showing the size of UCITS (Undertakings for the Collective Investment of
Transferable Securities) funds. These funds can be marketed within all EU countries irrespective
of their domiciliation.

Chart 5: European UCITS fund assets (for US and Japan mutual fund assets)

The chart shows tremendous growth of UCITS funds that are domiciled in Ireland and
Luxemburg. The euro area average amount of assets of this type of fund is about a third smaller
than the amount of mutual fund assets in the US. Easy access to mutual fund investments is an
important element in households’ portfolio management and therefore optimal consumption
behaviour.

Financial frictions are particularly relevant for the investment decision of small and young firms
(see for example equation (5)). These firms have no access to public capital markets and may
have difficulties obtaining private bank financing since they often have little collateral, no track
record and are usually high risk enterprises. A recent market form of financing these firms is
private equity or venture capital. Here, investors provide funds in return for an equity stake in the
firm or a debt claim that can be converted into equity at a later stage, and hope to recoup their
initial investment in an initial public offering. The advantage of this mode of financing is that in a
public offering the prospects of a high potential growth may overcome the constraints of having
limited collateral, a high business risk and strong information asymmetries. At the same time,
venture capital firms provide business advice and monitor firms in a way that is similar to banks.
Chart 6 shows the amount of venture capital financing that is spent at an early investment stage of
a project, while chart 7 shows the amount that is spent on expansions or replacements of existing
projects (both by country of management).

15
Charts 6: Venture capital financing (early investment stage) by country of management
Charts 7: Venture capital financing (expansion and replacement stages) by country of
management

The charts show that while venture capital is increasing the euro area, especially in Finland,
Sweden and Spain, it amounts only to a fifth of venture capital financing in the US and a half of
such financing in the UK. Among all the European countries, also the Netherlands have a more
active VC market, whereas this form of financing runs particularly low in Greece and perhaps
Austria and Italy.

Chart 8 shows the amount of venture capital financing (early investment and expansion) by
country of destination. The advantage of this indicator is that by focusing on the destination it is
more precise in measuring the amount of financing received by a specific country, the
disadvantage is that data for the United States are not comparable to Europe as they exclude
management buyouts. The results are roughly comparable to those in charts 6 and 7.

Charts 8: Venture capital financing by country of destination

3.3 Transparency and information

Transparency and information refer to measures that capture the degree to which a financial
system produces and spreads information about investment opportunities, market conditions and
the behaviour of economic agents. For example, the quality of accounting standards captures the
degree of asymmetric information between investors and managers. Better information on and
more transparent reporting of company performance alleviate the control problem between
outside investors and firm insiders, e.g. through more accurate stock prices, allowing outsiders to
exert pressure by selling their shares, or through an improved market for corporate control.

A measure of transparency and information that has been used in the finance and growth literature
comes from the disclosure of firms’ accounting information. Rajan and Zingales (1998) for

16
example use the CIFAR (Centre for International Financial Analysis and Research) index that is
based on the inclusion or omission of 90 items in firms’ annual reports. Chart 9 reports this index.

Chart 9: Accounting transparency (CIFAR index)

The quality of accounting information reported generally increases over time. For the latest data
available, the highest disclosure is found in Sweden, Finland, Ireland and the US. The smallest
number of accounting items is reported by firms in Portugal and Greece. 8

A potential shortcoming of such measures of disclosure though is that they do not capture
whether the information is actually used in capital markets. To address some of the issues, a
number of other measures of how a financial system deals with information frictions are
presented here. Chart 10 presents the extent of analyst coverage over time, while Chart 11 shows
the dispersion of analysts’ forecasts. Analysts study companies and make earnings forecasts.
They therefore represent an important outside assessment of a company’s condition. The more
companies are covered by analysts, the more information is available in a financial system.

Chart 10: Analyst coverage

In most countries, analyst coverage is either high or has increased significantly since 1990, except
in Germany where the extent of coverage has shrunk. Spanish, Greek and Irish companies are the
least covered companies in the euro area.

Related to the extent of analyst coverage is the dispersion of analysts’ forecasts. If a firm
discloses relevant and credible information, or if this information is readily available in a market,
then analysts’ earnings forecasts should converge. A financial system that aggregates and spreads

8
The latest data point available for the CIFAR index is 1995. Given changes in accounting rules in various
countries over time, it might therefore not reflect the most recent levels of firm transparency.

17
information more efficiently may therefore be associated not only with more analyst coverage,
but also with a lower dispersion of analysts’ forecasts.9

Chart 11: Dispersion of analysts’ forecasts

Chart 11 shows that analysts disagree most in Germany, Italy, Portugal, Sweden and Japan. The
extent of coverage in the euro area is comparable to that in the United States. The dispersion of
earnings forecasts, however, is more than twice as high.

The dissemination of information by stock markets is an important function of a financial system.


In order to determine how well a stock market incorporates useful information into stock prices
and how efficient it therefore is in guiding capital to the best investment opportunities, we
decompose stock price volatility into market volatility and firm-specific volatility. If firms’ stock
prices are mainly driven by market factors, i.e. if there is a high synchronicity among stocks, then
this is evidence that the stock market does not efficiently transmit firm-specific information. In
that case, the explanatory power (measured by the R2 statistics) of a regression of a firm’s stock
price on market factors should be high, i.e. most of firms’ stock price variation is explained
without firm-specific information (see for example Durnev et al (2003)).

To measure the information content of stock prices using the R2 statistic, we consider the
following model:

rijt = α + β 1 STMKT jt + β 2 EU t + β 3US t + β 4 EM t + ε ijt (6)

where rijt is the return on stock i in country j at time t, STMKTjt is the return on the stock market
index in country j at time t, and EUt, USt and EMt are the returns on the euro area, the US and on
an emerging market stock market indices at time t.

9
Analysts will also tend to agree less in periods of financial turmoil, e.g. Finland and Sweden in the early
1990s. This shows that economic and financial conditions may feed back into the indicators. Robustness
checks however show that the average dispersion over the last five years is uncorrelated with the average
stock market volatility over the same period but that it is negatively correlated with average GDP growth.

18
In this specification, stock prices are driven by two sets of explanatory variables. First, there are
the market factors (STMKT, EU, US, EM) that are common to all stocks in a market. They
capture market-wide information, i.e. the systemic information that enters into prices. Second,
there is the error term ε that captures the non-systemic or idiosyncratic element that drives stock
prices. It is assumed to pick up firm-specific news or events since they would not affect the
systemic factors. After computing R2 for a set of stocks in a given market, the average R2 is
computed for the market.

If the information disclosed by firms is relevant and credible, and if the stock market is efficient
in aggregating and spreading information, e.g. a great deal of informed trading takes place, then
the regression should not perform well in the sense that the R2 of the regression should be low. A
low R2 therefore indicates that the stock market is able to convey information about valuable
investment opportunities of firms (or the lack thereof). Conversely, if the R2 is high, the
explanatory power of systemic factors is high. In this case, stock prices move for reasons other
than firm-specific information, meaning that the stock market does not convey useful information
about individual firms’ investment opportunities.

The results reported in Chart 12 are based on a sample from 1990 to 2004 of 4,051 companies
listed in 17 countries.10

Chart 12: Pricing of firm-specific information in the stock market (R2 statistics)

Greece, Italy and Sweden have the highest R2, i.e. in these countries stock markets incorporate the
least amount of firm-specific information. The stock markets of Austria, Finland, Ireland and
Portugal have in the last few years incorporated a lot of firm-specific information into prices.
Since 2000, the euro area has incorporated on average more firm-specific information into stock
prices than the United Kingdom or the United States.

3.4. Corporate governance

10
The average country R2s from 2000 to 2004 are uncorrelated with average (i) industry composition and
sector concentration of firms, (ii) stock market volatility and (iii) GDP growth over the same period.

19
Corporate governance addresses the potential conflict between investors and managers, and
among investors, e.g. large versus small shareholders. Better governance ensures that their
interests are more aligned, that investors obtain a better return and that there will be a smaller loss
of efficiency due to opportunistic behaviour by managers.11

One example of good governance is strong shareholder protection. If for example investors are
allowed to vote in shareholder meetings even without being physically present, they can more
easily dismiss management. By threatening management with dismissal, shareholders will be
more likely to receive promised repayments, which in turn lowers the rate of return they demand
on their investments and the cost of capital.

Chart 13 presents a measure of how well the law protects minority shareholders in companies’
annual meetings (see La Porta et al (1997, 1998, 2000)). By this measure, shareholder rights have
improved in many European countries over the last years. In 2005 they were only slightly lower
in the euro area than in the US or the UK. Only the data for Belgium indicate very low
shareholder protection.

Chart 13: Shareholder rights

A recent alternative measure of shareholder protection quantifies their rights against expropriation
by corporate insiders through self-dealing (see Djankov et al. (2006)). Various forms of such self-
dealing include executive perquisites to excessive compensation, transfer pricing, self-serving
financial transactions such as directed equity issuance or personal loans to insiders, and outright
theft of corporate assets. These shareholder rights have improved in most European countries
(except Greece and Portugal) but are still not quite as strong as in the United Kingdom and the
United States.

Chart 14: Enforcement of shareholder rights against self-dealing (anti self-dealing index)
11
While the benefits of better governance for long-term growth are well understood, it may take time for
these benefits to materialise. Adopting better governance may have some adverse consequences in the short
run, as new laws and administrative measures may be burdensome at first. An open issue is to which extent
strong creditor rights could lead to more firm failures and to the unnecessary destruction of going-concern
value.

20
Chart 14 shows that the enforcement of shareholder rights against self-dealing by corporate
insiders is particularly easy in the UK, the US but also in Ireland (and to a lesser extent in
Belgium and Portugal). They are much weaker in Austria, Greece, Luxemburg and the
Netherlands. The euro area average score on this index of shareholder protection is about a half of
the score for the US and about a third of the score for the UK.

A related but more difficult measure of corporate governance is the protection of creditors’ rights.
Different kinds of creditors have different interest, e.g. senior versus junior creditors. Also, a
defaulting firm may either be reorganised or liquidated which makes it difficult to establish when
exactly creditors are well protected. We follow the methodology of La Porta et al (1997, 1998,
2000) and present, and update, in chart 15 a measure that i) focuses on senior creditors and ii)
does not take a stand on whether reorganisation or liquidity is more efficient.

Chart 15: Creditor rights

Creditor rights are weakest in Finland, Greece, Ireland, Portugal, Sweden, Switzerland and also
the US. The UK has by far the strongest protection of creditors. The rights of Japanese creditors
appear to have decreased recently.

The classic image of the modern corporation run by professional managers and dispersed small
outside shareholders has been questioned by La Porta et al. (1998, 1999). They show that except
in the US, firms tend not to be widely held and most corporations have outside shareholders with
significant stakes. Chart 16 presents the shareholding of the largest and the ten largest
shareholders for the ten largest publicly traded firms in our set of euro area and benchmark
countries.

Chart 16: Ownership concentration (size of shareholdings)

21
We see that firms in the euro area have on average a largest shareholder who holds a stake that is
more than three times as large as his counter part in the US, and that is more than five times as in
the UK. The largest shareholder is most dominant in Italy and Portugal.12 The ten largest
shareholders together hold 43% of the largest euro area companies, and they hold around 27% in
the US and the UK.

As an alternative measure of ownership concentration, Chart 17 shows the Herfindahl index of


the shareholdings of the largest twenty shareholders. Most euro area countries have comparable
shareholder concentration at a level that is more than five times as large as in the US, and
especially the UK. Ireland and the Netherlands, and to a lesser extent, Sweden and Switzerland
have somewhat lower concentration.

Chart 17: Ownership concentration (Herfindahl index)

The academic literature has identified cost and benefits for a firm of having large outside
shareholders. On the one hand, having a large outside shareholder facilitates take-overs and
monitoring both of which exerts positive pressure on management. On the other hand, large
outside shareholders may derive private benefits of control and therefore not act in the interest of
minority shareholders (Shleifer and Vishny (1986)). La Porta et (1999, 2000) argue that the cost
of having concentrated ownership outweighs its benefits. Countries with a better protection of
minority shareholders tend to have more dispersed ownership. The same holds in our sample of
industrialised countries (compare Charts 13 and 14 with Charts 16 and 17). Better legal protection
means that shareholders can give up control, reduce their equity stake and diversify more, which
in turn leads to more liquid stock markets.

The identity and type of dominant outside shareholders is also relevant for corporate governance.
Research has shown that i) having a large outside shareholder is detrimental if that shareholder is
a family or another entrenched individual or entity (see Morck et al (2005) for a survey of this
rapidly growing literature together with implications for growth), and that ii) institutional
investors actively intervene to improve the corporate governance of firms whose shares they hold

12
Luxemburg is an outlier since the high concentration there is due to the RTL group that is wholly owned
by Bertelsmann, a German media company.

22
(Hartzell and Starks (2003)). Chart 18 shows the proportion of large outside shareholders that are
institutions.

Chart 18: Proportion of institutions among largest shareholder

While 9 out of the 10 largest US or UK firm do have an institution (defined here as an entity in
the business of investment management) as the largest shareholder, the same is true on average in
the euro area for not even 2 out of the 10 largest firms.

3.5. Legal system

In the perfect market benchmark, there are no frictions because contracts covering all possible
future contingencies can be written and, equally importantly, can be enforced. When a financial
system allocates capital across time and space, contracts are needed to connect providers and
users of capital. A financial investor relinquishes control of his funds now in return for a
promised claim to future cash flows. In order for the promise not to be an empty one, contracts
are written and a well-functioning financial system must have an implicit or explicit mechanism
to enforce them. The legal system, among other things, explicitly enforces financial contracts and
thus contributes to the performance of financial systems.

It is extremely difficult to assess the many facets of a legal system in relation to the financial
system and many different measures of legal effectiveness are conceivable. Chart 19 presents one
possible measure that covers adherence to the rule of law: the “Law and Order” index. It has been
used successfully in the legal and financial literature (see La Porta et al (1998), Erv et al (1996) or
the United Nations “Human Development Report”). The index intends to capture the strength and
impartiality of a legal system as well as popular observance of the law.

Chart 19: Law and Order index

23
Due to the limited scope of the index, its results must be interpreted very carefully. It does not for
example consider a legal system’s procedures and arrangements or its ability to foresee conflicts.
According to the Law and Order index, all the countries considered here score relatively highly
(except for Greece recently), as one would expect for most industrial countries. On average, the
level of the Law and Order index for the euro area is comparable to that for the United States or
Japan.

Chart 20 through 22 report indicators that measures in more detail how well a legal system solves
financial conflicts and enforces laws that protect investors. Recent research shows that not only
the existence of such laws but also their enforcement reduces frictions impeding the flow of
capital (see for example Bhattacharya and Daouk (2002), Djankov et al (2003) and La Porta et al
(2006)). The indicator in Chart 20 asks how many days it takes on average in a country to recoup
a bounced cheque through the courts. A speedy resolution of financial conflict in courts reduces
administrative costs and enhances the trust in financial contracts, thereby improving the
functioning of a financial system.

Chart 20: Duration of enforcement

While the courts in most euro area countries allow a quick recovery of bounced cheques, it takes
considerably longer to do so in Austria, Portugal and especially Italy.13 It takes the least amount
of time to resolve this particular type of financial conflict in the Netherlands and in Japan. The
average time for the recovery of bounced cheques in the euro area is comparable to that in the
United Kingdom and the United States.

Chart 21 shows the degree to which managers believe the legal system will uphold contracts and
property rights in a business dispute. Should those rights be ill protected then there will be serious
frictions to the allocation of capital from investors to firms and their managers.

Chart 21: Enforcement of property rights

13
The 2004 data are not exactly comparable with the 2003 data (see the notes of the chart for more details).

24
On average, managers from euro area countries believe that the property rights are worse
protected than managers in the US or the UK. Within the euro area, Spain, France, Greece, Italy
and Portugal appear to have weaker protection of property rights.

Table 2 compares the existence and the enforcement of insider trading laws across countries.
Bhattacharya and Daouk (2002) show that the start of enforcement of insider trading laws, as
indicated by the first prosecution under it, is associated with a drop in the cost of equity for firms.
All the countries in our sample do have insider trading laws, with the US being the first country to
adopt them. There are some differences in when these laws were enforced. The earliest
enforcements took place in the US, followed by France and the UK. Spain, Portugal and
Luxemburg were the last countries in the sample to enforce such laws.

Table 2: Laws on insider trading

3.6. Regulation and supervision of the financial sector and financial stability

It is widely recognised that the financial sector is “special” compared with many other sectors of
the economy. First, it faces a greater risk of instability, both at the level of individual financial
intermediaries and markets and at the level of the overall financial system. In particular, systemic
financial crises can have large adverse effects on growth. Second, many households using retail
financial services may lack financial knowledge and the ability to collect information about the
nature and risks of various financial contracts and about the viability of financial intermediaries to
whom they entrust their savings. For these reasons, financial sectors tend to be subjected to more
regulation and supervision than most other sectors (see Dewatripont and Tirole (1994) or
Goodhart et al. (1998)). This regulation and supervision aims to stabilise financial intermediaries
and financial systems as well as to protect consumers. In pursuing financial stability and
consumer protection, efficiency can however sometimes suffer, for example when regulations
inadvertently deter efforts to innovate, lead to adverse risk-shifting incentives or impose
excessive administrative costs on financial intermediaries (see Shleifer and Vishny (1998)). We
follow and update the analysis of Barth et al (2004) who examine bank regulation and supervision
in a large number of countries.

25
Chart 22 shows an index of how much power the bank supervision authority has in a given
country, i.e. whether the supervisor can intervene in, restructure or declare the insolvency of
banks. Banks are costly and difficult to monitor so that a power supervisor can step in and fill this
gap. But a powerful supervisor can also use its power to serve his own or vested interest, e.g. to
favour certain political constituents. The most powerful supervision authorities are found in
Luxemburg, Switzerland and the US, the weakest ones in France, Italy and perhaps the
Netherlands.

Chart 22: Bank supervisor power

Chart 23 presents a measure of whether the supervisor can engage in forbearance when banks
violate regulations or act imprudently and chart 24 measures the degree to which the supervisory
authority is independent from political pressure and protected from lawsuits by supervised banks.
While too much forbearance is regarded as detrimental, since it creates moral hazard on the part
of banks, supervisory independence should be beneficial for the efficiency of the banking system.

Chart 23: Supervisory forbearance discretion

Chart 24: Bank supervisor independence

The figure shows that there are significant differences across countries, with some benefiting
from a lot of discretion (e.g. Germany, Sweden, the UK and perhaps Belgium) and others not
(including Finland, Spain, Japan and the US). On average there appears to be more discretion for
supervisory forbearance in Europe than in the US. This is to a large extent a consequence of the
US Federal Deposit Insurance Improvement Act introduced in the 1990s, which limits the risk of
moral hazard by prompt corrective action provisions. On the other hand, the supervisors are more
independent in the euro area than in the US (with the exception of Sweden).

Deposit insurance scheme are an attempt to prevent bank runs by depositors. But such schemes
come at a cost of increasing the moral hazard of banks. Knowing that government will guarantee
the claims of their depositors, banks may take excessive risks and therefore destabilise the
financial system. Chart 25 presents an index used by Demirgüc-Kunt and Detragiache (2002) to

26
show that more extensive deposit insurance, e.g. schemes that cover wholesale deposits and are
fully funded, increases the likelihood of a costly financial crisis.

Chart 25: Moral hazard index of deposit insurance

Chart 25 shows again significant differences across countries. High bars, indicating a significant
risk of moral hazard, are found for Finland, Germany, Italy and the US. In contrast, the risk of
bank moral hazard due to deposit insurance seems to be low in Switzerland.

Similarly, restrictions on bank activities could, on the one hand, make banks more transparent,
and thus easier to monitor, and limit their degree of market power. On the other hand, restrictions
could prevent banks from exploiting economies of scale and limit their franchise value, thereby
reducing their incentives for sound and prudent behaviour. Chart 26 shows whether banks are
allowed to undertake fee-based activities, e.g. the underwriting and selling of securities and
insurance, in addition to deposit-taking and lending.

Chart 26: Bank activity restrictions

Most European banks were previously unrestricted in their activities, but are now not allowed to
undertake the full range of activities in the insurance business (except in France and Ireland). In
contrast, US and Japanese banks were strongly regulated, but recently saw a loosening of the
restrictions across both securities and insurance businesses.

Supervisors can also encourage private monitoring by capital markets to complement their own
effort. Recent regulatory efforts place indeed increasing importance on accurately assessing risks
and on the role of market discipline.14 Chart 27 therefore presents a measure that relies on
potential complementarities between regulation and private market monitoring. It aggregates
formal regulations that ease the private monitoring of banks, in particular by wholesale investors,

14
See, for example, the provisions in pillar 3 of the new framework of the Basel Committee on Banking
Supervision (“Basel II: International Convergence of Capital Measurement and Capital Standards: a
Revised Framework”, June 2004).

27
e.g. accounting and audit requirements. Stronger incentives for private bank monitoring in turn
have been shown to lower net interest margins and to reduce the proportion of non-performing
loans.

Chart 27: Bank regulations supporting market discipline

Taken together, the indicators of banking regulation in Charts 22 to 27 present a picture of


supervision in the euro area that is comparable to supervision in the US and UK. At the same
time, there is a considerable degree of heterogeneity across measures and countries in the euro
area.

While most analysis of regulation and supervision of intermediaries in financial markets focuses
on banks, insurance companies are heavily regulated as well.15 Chart 28 quantifies the power of
the regulator in the insurance sector. For example, can the regulator require insurers to comply
with prescribed solvency regimes (e.g. capital requirement), carry out on-site inspections, enforce
corrective measures and impose sanctions?

Chart 28 Insurance supervisor power

Chart 29 captures regulatory transparency. This measure includes both the extent to which the
regulator’s authority and objective are clearly defined and its actions carried out in a transparent
fashion, as well as the degree to which insurance companies are forced to disclose relevant
information on a timely basis in order to give stakeholders a clear view of the risks to which they
are exposed.

Chart 29: Transparency of insurance sector regulation

15
To our knowledge there has been no published academic research on supervision in insurance. The
reason appears to be a lack of data.

28
Overall, there seems to be very little difference among the countries for which we have data in
how the insurance sector is regulated (except for the Netherlands with respect to transparency).

3.7. Competition, openness and financial integration

More competition among suppliers of capital in a financial system reduces frictions. It eliminates
inefficient suppliers, frees up resources that are captured through market power and ensures value
maximisation as a means of survival. Less restrictions, openness and integration in turn support
competition by easing the exchange of goods and services and allowing easier entry of
competitors.16

However, with respect to banking, the overall effect of competition on growth is theoretically
ambiguous. More competition could force banks to lower lending rates and increase the provision
of credit. But it could also decrease incentives to acquire information about borrowers and to
monitor them, leading to poorer loan quality and a higher cost of capital. In practice, however, the
benefits of banking competition appear to outweigh its costs (Claessens and Laeven (2005)).

It is very difficult to assess the competitiveness of a market. One standard but imperfect measure
of competition is market concentration. The more concentrated a product-differentiated market,
such as a loan market, is the greater may be monopoly profits and the higher loan rates. Charts 30
shows the Herfindahl index based on banks’ assets.

Chart 30: Bank concentration

A number of European countries have reached a high level of bank concentration, notably
Austria, Belgium, Finland, the Netherlands, Switzerland and the UK. This contrasts with the
situations in Germany, France, Italy, Luxembourg and Spain as well as with those in Japan and

16
See, for instance, London Economics (2002), “Quantification of the macroeconomic impact of integration
of EU financial markets”, a study commissioned by the European Commission on the economic benefits of
financial integration, which is available on the Commission’s website. Moreover, Giannetti et al (2002)
argue that integration and openness supports further domestic financial development through external
competitive pressures.

29
the US where domestic consolidation was more limited so far or did not have the same
concentrating effect.17 Apart from challenges concentration poses for competition, there may also
be an enhanced risk of political involvement in the banking sector.

One possible way to counter domestic bank concentration is via foreign bank entry. As a general
rule, however, foreign bank penetration tends to be quite low among industrial countries. Chart 31
depicts the asset market shares of foreign bank branches and subsidiaries and confirms that except
for Luxembourg and recently Finland (driven by changes in the company structure of Nordea),
foreign penetration is very limited in all sample countries.

Chart 31: Foreign bank penetration

Chart 32 through 36 present alternative measures of competition in banking using balance sheet
and income statement information. Competitive forces should erode profits and foster efficiency.
But even under those forces, banks may continue to make considerable profits since there are
economies of scale and fixed costs involved in lending and deposit taking (see for example
Diamond (1984)).

Chart 32: Net interest income over assets

Chart 33: Net interest margins (for commercial banks)

Charts 32 and 33 show that US and Spanish banks have the highest interest income and interest
margins, while banks in Luxemburg have the lowest. Overall, banks in the euro area have
somewhat less interest income and interest margin than banks in the US but are comparable to
banks iin the UK. A different picture emerges by looking also at non-interest income in Chart 34.
First, there appears to be a trend away form interest to non-interest incomes, i.e. less lending and
more trading and fee based activities such as underwriting or origination of securities. Second,
US and UK banks obtain much more income from these activities than banks in the euro area.
And third, this trend towards non-interest income is stronger in the US than in Europe (see also

17
In Germany concentration would be somewhat higher, if savings and co-operative bank networks were
regarded as a single entity.

30
Decressin et al (2004)). An inability to develop significant non-interest business could be a sign
of weakness, as it is likely that the banking sector will be moving even further away from
traditional forms of intermediation in the future.

Chart 34: Net non-interest income over assets

But although euro area banks obtain less income than banks in the US, they also have lower
operating expenses (Chart 35). Overall then banks cost to income ratios are quite similar across
countries and over time at a level of just above or below 60% (see Chart 36). Some exceptions are
Finland and Sweden in the early 1990s and Japan since 2000. Banking in Luxemburg appears to
be more efficient than elsewhere.

Chart 35: Operating expenses over assets

Chart 36: Cost to income ratio

Despite the similarity in cost-income ratios, banks’ return on assets vary significantly across
countries and over time (see Chart 37). The returns are lowest in Germany (except for Japan
where returns are negative since 1995), likely related to the inability of German banks to move
successfully away from traditional interest-rate based to more innovative fee-based business (see
Decressin et al. (2004)), and highest in the US and Finland (since 2000).

Chart 37: Return on assets

A more structural though controversial approach to measuring competition in banking is to


examine how banks’ revenues vary with changes in factor input prices (see Rosse and Panzar
(1977) and, more recently, Claessens and Laeven (2004)). Under competition, increases in input
prices lead to both higher marginal costs and higher revenues, while the latter decreases in a
monopoly. With this insight and making a number of simplifying assumptions, one can measure
competition with the following reduced form revenue equation:

31
ln REVit = α + β 1 ln INTEXPit + β 2 ln LABOURit + β 3 ln OTHERit + CONTROLS + ε it (8)

where REV is interest income over total assets, INTEXP is interest expenses over deposits and
money market funding, LABOUR is personnel expenses over assets, OTHER is other operating
and administrative expenses over assets, CONTROLS are book equity over assets, loans over
assets, assets, operating income minus net interest income over assets and time dummies; i
indexes banks and t indexes time. Using the estimated coefficients from regression (8), we can
compute the so-called “H-Statistic”, H = β1 + β 2 + β 3 . When H=1 the market is assumed to be
competitive since the elasticity of revenue with respect to costs is one. As markets become less
competitive, this elasticity decreases since there are less competitive forces that pressure banks
into adjusting revenues to costs.

Chart 38: Bank competition (H-Statistic)

According to the H-statistic, the Dutch, and Luxembourg (but also Greek) banking sector are the
most competitive, while Japanese banking is the least competitive. According to this measure,
euro area banking is on average more competitive than US or UK banking. Some of these results
seem rather counter-intuitive and may suggest that the restrictive assumption of the H-statistic do
not hold in a number of countries.

Finally, we present a measure of state ownership of banks (Chart 35). Higher government
ownership of banks limits competition and entry, and it is associated with lower financial
development and lower economic growth (La Porta et al (2002)). Research shows that
government owned and controlled banks may pursue political objectives (e.g. lending to
politically connected firms and entrepreneurs) at the expense of profit maximisation (see for
example Dinc (2005) or Sapienza (2004)).

Chart 39: State ownership of banks

32
Significant state ownership of banks is still present in some countries of the euro area, especially
in Germany but also in Portugal and Greece.18 There is no state ownership of banks in the UK, the
US or Spain.

3.8. Economic freedom, and political and socio-economic factors

The functions of a financial system centre on the provision of information, the enforcement of
contracts and the facilitation of transactions. These functions are affected by political and socio-
economic factors in addition to the more tangible forces presented so far. This last section
mentions some issues that are harder to measure but nevertheless are important for a broad
discussion of the performance of financial systems.

Not only the legal system but also the general institutional environment partly determines the
functioning of a financial system (see Williamson (2000) for a conceptual framework).
Governance structures that mitigate the control problem between financial investors and
managers are, for example, embedded in traditions, social norms, religion and politics. A
manager’s sense of duty towards outside investors will ease the conflict of interest between him
and his investors even when there are few formal governance arrangements. The idea is that
explicit contracts can neither anticipate nor include all possible contingencies, nor can they be
perfectly enforced under all circumstances. All economic exchange depends to some extent on
trust and fairness, and on what is perceived as “fair”, which are all shaped by socio-economic and
ethical factors. This is particularly pronounced in the financial sector, where contracts have a
strong inter-temporal component and providers of funds have to rely on users of funds being able
and willing to return the money in the future. It has been shown that social capital, measured
either using surveys on how people trust each other or using a metric of civic engagement such as
voter turnout at local elections, matters in a financial system (see for example Guiso et al (2004)).
Similarly, politics shape the laws governing creditor and shareholder protection and partly
determine the protection of private property rights vis-à-vis the rights of the State.

18
In Portugal, one large bank accounts for most of the state ownership in banking, whereas in Germany it is
mainly due to the large network of savings banks.

33
Economic freedom captures the notion that fewer administrative burdens on economic activity,
e.g. less red tape, means less frictions in the flow of capital. The indicator of economic freedom
shown in Chart 40 measures “the absence of government coercion or constraint on the production,
distribution, or consumption of goods and services beyond the extent necessary for citizens to
protect and maintain liberty itself.” (Miles et al (2006)).

Chart 40: Economic freedom

The indicator of economic freedom does not show large differences across countries. However, it
is noteworthy that economic freedom has increased in all countries since 1995 except Italy, Japan
and to some extent Greece, which also has the lowest economic freedom. The UK in contrast has
the highest economic freedom according to this indicator.

Related to government involvement in a countries economy is the starker issue of corruption in


government. Chart 41 presents a measure that has been used successfully in the finance and
growth literature (for example by La Porta et al (1997, 1998) and Djankov et al (2004)).

Chart 41: Control of corruption

While the euro area level of control of corruption is comparable to the one in the UK and the US,
there are some differences within Europe. Finland, Swede, Switzerland and Luxemburg achieve
the best scores while Italy and Greece have the lowest scores in the sample of countries.

4. Tentative conclusions

This work presents a framework for assessing the performance of financial systems in developed
countries. The approach taken is a functional one, i.e. a financial system is viewed as performing
a number of functions to overcome market imperfections. The article shows how the functional
framework translates into a number of groups of financial system characteristics that can be used
to structure the assessment of performance. The quality with which a financial system performs

34
its functions can then be evaluated with a set of economic indicators under each group. The
framework and the choice of indicators are particularly geared towards the most important
dimensions for industrialised countries with relatively developed financial systems.

The article applies a selection of these indicators to euro area countries and to a reference group
of non euro area countries. Some are updates of indicators used previously in the finance and
growth literature and others are new. A number of preliminary conclusions emerge from them.
First, there is in general a fair amount of heterogeneity in financial system performance across
euro area countries. Second, a country’s performance, as measured by the indicators displayed,
can vary a lot across functions and dimensions. Some countries that score high in one dimension
of a financial system may not necessarily do so in another.

Whereas the financial systems of many European countries have improved over time, there is still
further room for financial development or modernisation. Our indicators suggest that Greece,
Italy and Portugal appear to have more room for further improvements of their financial systems,
while Finland, the Netherlands and Sweden are front-runners in terms of financial development.
The large financial systems of France and Germany offer a somewhat mixed picture. The UK,
and to a lesser extent the US, somewhat stand out as performing better across a large number of
indicators than euro area countries. The large differences in performance identified so far seem to
suggest that there is further room for structural reforms of financial sectors in the euro area. These
reforms have a high potential to foster productivity and economic growth in Europe.

35
References

Allen F. and D. Gale, 1997, “Financial Market, Intermediaries and Intertemporal Smoothing”,
Journal of Political Economy, 105, 523-546.

Allen and Gale (2000), Comparing Financial Systems (Cambridge, MA: MIT Press).

Barth, Caprio and Levine (2004), Bank regulation and supervision: What works best?, Journal of
Financial Intermediation

Bhattacharya, U., and H. Daouk, 2002, “The World Price of Insider Trading”, Journal of
Finance, 57, 75-108.

Claessens and Laeven (2004), What drives bank competition: Some international evidence,
Journal of Money Credit and Banking, Vol. 36.

Claessens and Laeven (2005), Financial sector competition, financial dependence and growth,
Journal of the European Economic Association, 3 (1)

Coase (1937), The nature of the firm, Economica.

Decressin, Brunner, Daniel and Kudela (2004), Germany’s three pillar banking system: cross-
country perspectives in Europe, IMF Occasional Paper No 233

DeMarzo, 2005, The Pooling and Tranching of Securities: A Model of Informed Intermediation,
Review of Financial Studies, 18, 1-35.

Dewatripont, M., and J. Tirole, 1993, The prudential regulation of banks, (Cambridge, MA: MIT
Press).

Diamond (1984), Financial intermediation and delegated monitoring, Review of Economic


Studies, 51, 393-414.

Diamond and Dybvig (1983), Bank runs, deposit insurance and liquidity, Journal of Political
Economy, 91, 401-419.

Demirgüc-Kunt and Detragiache (2002), Does deposit insurance increase banking system
stability? An empirical investigation, Journal of Monetary Economics.

36
Demirgüc-Kunt and Levine (eds., 2001), Financial Structure and Economic Growth (Cambridge,
MA: MIT Press).

Dinc, S., 2005, “Politicians and Banks: Political Influences on Government-Owned Banks in
Emerging Countries”, Journal of Financial Economics, 77, 453-479.

Djankov, La Porta, López-de-Silanes and Shleifer (2002), “The Regulation of Labor”, Quarterly
Journal of Economics.

Djankov, La Porta, López-de-Silanes and Shleifer (2003), Courts: The Lex Mundi Project,
Quarterly Journal of Economics, 65, pp. 453-517.

Djankov, S., C. McLiesh, and A. Shleifer, 2006a, “Private Credit in 129 Countries”, Journal of
Financial Economics, forthcoming.

Djankov, S., R. La Porta, F. Lopez-de-Silanes, and A. Shleifer, 2006b, “The Law and Economics
of Self-dealing”, working paper.

Durnev, Morck and Yeung (2003), “Value enhancing capital budgeting and firm-specific stock
returns variation”, Journal of Finance, 59 (1), pp. 65-106.

Edison, Levine, Ricci and Slok (2002), International financial integration and economic growth,
Journal of International Money and Finance.

Erv, Harvey and Viskanta (1996), Political risk, economic risk and financial risk, Financial
Analysts Journal, vol. 52 (6)

Gaspar, Hartmann and Sleijpen, eds (2002), The Transformation of the European Financial
System (Frankfurt: European Central Bank).

Giannetti, Guiso, Japelli, Padula and Pagano (2002), Financial integration, corporate finance and
economic growth, European Economy Economic Papers, No. 179.

Goodhart, Hartmann, Llewellyn, Rojas-Suarez and Weisbrod (1998), Financial Regulation: Why,
How and Where Now? (London: Routledge).

Guiso, Sapienza and Zingales (2004), The role of social capital in financial development,
American Economic Review, 94 (3), pp. 526-56.

37
Hartmann, Maddaloni and Manganelli (2003), The euro-area financial system: Structure,
integration and policy initiatives, Oxford Review of Economic Policy.

Hartzell and Starks (2003), Institutional investors and executive compensation, Journal of
Finance.

Jensen, M.C., and W.H. Meckling, 1976, “Theory of the firm: Managerial behavior, agency costs
and ownership structure”, Journal of Financial Economics, 3, 303-360.

Jorgenson (1995), Productivity, Vol. I and II (Cambridge, MA: MIT Press).

Jorgenson (2005), Growth accounting, in P. Aghion and S. Durlauf (eds.), Handbook of Economic
Growth (Amsterdam: North Holland).

Kaplan and Zingales (1997), Do investment-cash flow sensitivities provide useful measures of
financing constraints?, Quarterly Journal of Economics.

King and Levine (1993), Finance and Growth: Schumpeter Might Be Right, Quarterly Journal of
Economics, 153, 717-738.

Levhari and Srinivasan (1969), Optimal savings under uncertainty, Review of Economic Studies.

Levine (2002), “Bank-based or market-based financial systems: which is better?” Journal of


Financial Intermediation, 11, pp. 398-428.

Levine (2005), Finance and growth: Theory and evidence, in P. Aghion and S. Durlauf:
Handbook of Economic Growth, Elsevier.

Levine and Zervos (1998), Stock markets, banks and economic growth, American Economic
Review

La Porta, López-de-Silanes, Shleifer and Vishny (1997), Legal determinants of external finance,
Journal of Finance.

La Porta, López-de-Silanes, Shleifer and Vishny (1998), Law and Finance, Journal of Political
Economy.

La Porta, López-de-Silanes, Shleifer and Vishny (1999), Corporate ownership around the world,
Journal of Finance.

38
La Porta, López-de-Silanes, Shleifer and Vishny (2000), Investor protection and corporate
governance, Journal of Financial Economics, vol. 58 (1-2), pp. 3-27

La Porta, López-de-Silanes, Shleifer and Vishny (2002), Government ownership of banks,


Journal of Finance.

La Porta, R., F. Lopez-De-Silanes, and A. Shleifer, 2006, “What Works in Securities Laws?”,
Journal of Finance, 61, 1-32.

Miles, O'Grady and Holmes (2006), The 2006 Index of Economic Freedom, The Heritage
Foundation, (www.heritage.org/research/features/index)

Modigliani and Miller (1958), The cost of capital, corporation finance and the theory of
investment, American Economic Review.

Morck, Wolfenzon and Yeung (2005), Corporate governance, economic entrenchment and
growth, Journal of Economic Literature.

Myers, S., and N. Majluf, 1984, “Corporate Financing and Investment Decisions when Firms
have Information that Investors Do Not Have”, Journal of Financial Economics, 13, 187-
221.

Rajan and Zingales (1998), Financial dependence and growth, American Economic Review,
88 (3), pp. 559-86.

Rosse and Panzar (1977), Chamerlin vs Robinson: an Empirical Test for Monopoly Rents, Bell
Laboratories Economics Discussion Paper, No 90.

Sapienza, P., 2004, “The Effects of Government Ownership on Bank Lending”, Journal of
Financial Economics, 72, 357-384.

Shleifer and Vishny (1986), Large shareholder and corporate control, Journal of Political
Economy.

Shleifer and Vishny (1998), The grabbing hand: government pathologies and their cures,
(Cambridge, MA: Harvard University Press).

Stein (2001), Agency, information and corporate investment, Handbook of Finance.

Stiglitz, J.E., and A. Weiss, 1981, “Credit Rationing in Markets with Imperfect Information”,
American Economic Review, 71, 393-410.

Williamson (2000), The New Institutional Economics: Taking Stock, Looking Ahead, American
Economic Review, vol. 38, no 3, pp. 595-613.

39
Chart 1: Size of capital markets
(percentages of GDP)
1990-1994 1995-1999 2000-2005
500%

450%

400%

350%

300%

250%

200%

150%

100%

50%

0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: BIS, Eurostat, IMF International Financial Statistics, World Federation of Exchanges and ECB
calculations.
Notes: Sum of (i) stock market capitalisation, (ii) bank credit to the private sector and (iii) domestic debt securities
issued by the private sector divided by GDP. Data for Luxembourg exclude debt securities. Euro area (EA) figures are
averages of EA country data weighted by GDP.

Chart 2: Turnover ratio


(percentages of market capitalisation)

1990-1994 1995-1999 2000-2005


4.0

3.5

3.0

2.5

2.0

1.5

1.0

0.5

0.0
UK
LU

CH
GR
AT

DE

NL
FR
BE

IE

IT

PT

SE

US
ES

JP
FI

Euronext

Data sources: World Federation of Exchanges, Datastream and ECB calculations.


Notes: turnover is defined as trading volumes divided by market capitalisation. Data for BE, FR and NL are from 1991
until 2000; for PT from 1993 until 2001. Data for Euronext start in 2001, for ES in 2002, for IE in 1996 and for CH in
1991. Turnover for US is the Nasdaq turnover.

40
Chart 3: Securitisation
(percentages of GDP, by country of collateral)
2000 2002 2005

25%

20%

15%

10%

5%

0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK

Data sources: European Securitisation Forum, Eurostat and ECB calculations.


Notes: Issuance placed in the Euromarket or in European domestic markets. Data include asset-backed securities,
mortgage-backed securities and Pfandbriefe. No data for US, JP and FI (in 2005). Euro area (EA) figures are averages
of EA country data weighted by GDP.

Chart 4: Turnover of interest rate derivatives


(percentages of GDP)
1995 1998 2001 2004

30%

25%

20%

15%

10%

5%

0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: BIS, IMF and ECB calculations.


Notes: Daily average turnover in the month of April of OTC single currency interest rate derivatives (net of local inter-
dealer double-counting). Euro area (EA) figures are averages of EA country data weighted by GDP.

41
Chart 5: European UCITS fund assets (for US and Japan: mutual fund assets)
(percentages of GDP by domicile nation of a fund)
2000 2002 2005

300%
LU
2000: 3600%
250% 2002: 3190%
2005: 4730%

200%

150%

100%

50%

0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: EFAMA, Eurostat and ECB calculations.


Notes: Data are net assets of nationally domiciled UCITS (Undertaking for Collective Investment in Transferable
Securities). That means that the funds that are domiciled abroad and promoted by national sponsors are not counted in
the sponsor’s country but in the country where they are domiciled. Fund of funds are included in figures for IT, FR
and LU. Euro area (EA) figures are averages of EA country data weighted by GDP.

Chart 6: Venture capital financing (early investment stage) by country of management


(percentages of GDP, by country of management)
1995-1999 2000-2004
0.15%

0.10%

0.05%

0.00%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: European Private Equity and Venture Capital Association, PricewaterhouseCoopers and Eurostat.
Notes: Venture capital investment is defined as private equity raised for investment in companies; management
buyouts, management buyins and venture purchases of quoted shares are excluded. Early investment stage means seed
and start-up financing. Data report venture capital funds raised from companies in each country. No data for LU and
JP. CH data start in 1999. US data end in 2002. Euro area (EA) figures are averages of EA country data weighted by
GDP.

42
Chart 7: Venture Capital financing (expansion and replacement stages), by country of
management.
(percentages of GDP)
1995-1999 2000-2004

0.50%

0.40%

0.30%

0.20%

0.10%

0.00%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: European Private Equity and Venture Capital Association, PricewaterhouseCoopers and Eurostat.
Notes: Venture capital investment is defined as private equity raised for investment in companies; management buyouts,
management buyins and venture purchase of quoted shares are excluded. Expansion and replacement stages means funds
used to expand and replace capital invested at early stage. Data report venture capital funds raised from companies in
each country. No data for LU and JP. CH data start in 1999. US data end in 2002. Euro area (EA) figures are averages of
EA country data weighted by GDP.

Chart 8: Venture Capital financing by country of destination


(percentages of GDP)
1 999 2002 200 5

1.0%

0.8%

0.6%

0.4%

0.2%

0.0%
AT BE DE ES FI FR GR IE LU IT NL PT EA CH S E UK JP US

Data sources: European Private Equity and Venture Capital Association, National Venture Capital Association,
Eurostat and ECB calculations.
Notes: Data for European countries come from a survey conducted by EVCA that focuses on all European Private
Equity funds acting as General Partners (GP). A private Equity GP is a fund that raises money from a third party called
Local Partner (LP) (banks, insurance companies, pension funds, private individuals etc...) and invests in private
companies. Data include VC investments and buyouts. No Data for LU and JP. Data for the US do not include buyouts.
Euro area (EA) figures are averages of EA country data weighted by GDP.

43
Chart 9:Accounting transparency
(CIFAR index)
1990 1993 1995

90

80

70

60

50

40

30

20

10

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: CIFAR (Center for International Financial Analysis and Research) and ECB calculations.
Notes: the index is a measure of mandatory and voluntary disclosures made by firms. It has been created by examining
and rating companies’ annual reports on their inclusion or omission of a set of selected items. It ranges from 0 to 90. No
Data for LU. Euro area (EA) figures are averages of EA country data weighted by GDP.

Chart 10: Analyst coverage


(percentages of total stock market capitalisation)
1990-1994 1995-1999 2000-2004

100%

90%
80%

70%

60%

50%

40%
30%

20%

10%

0%
AT BE DE ES FI FR GR IE IT NL PT EA CH SE UK JP US

Data sources: Thomson Financial’s First Call database, IMF and ECB calculations.
Notes: Stock market capitalisation of firms covered by at least one forecast of the earnings per share for current fiscal
year divided by total stock market capitalisation. Data for LU are not available, while for the NL and PT the data end in
2001. Euro area (EA) figures are averages of EA country data weighted by the stock market capitalisation of each
country.

44
Chart 11: Dispersion of analysts’ forecasts
(percentages of forecasted earnings per share)
1990-1994 1995-1999 2000-2004

1.2

1.0

0.8

0.6

0.4

0.2

0.0
AT BE DE ES FI FR GR IE IT NL PT EA CH SE UK JP US

Data sources: Thomson Financial’s First Call database and ECB calculations.
Notes: Average standard deviation of the earnings per share (EPS) forecasts for a given year divided by the level of the
EPS forecasts for that year. The average standard deviation of the EPS is an average of firm level standard deviation of
EPS forecasts weighted by the market capitalisation of firms and the level EPS forecast is an average of the firm level
EPS forecast weighted by the market capitalisation of firms. Data for LU are not available, while for the NL and PT
the data end in 2001. Euro area (EA) figures are averages of euro area country data weighted by the stock market
capitalisation covered by analysts in each country.

Chart 12: Pricing of firm-specific information (R2 statistics)


1990-1994 1995-1999 2000-2004

0.4

0.3

0.2

0.1

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Datastream and ECB calculations.


Notes: Country average R2 statistics for the regressions described in section 3.3. Low bars indicate that stock prices
reflect more firm specific information. Euro area (EA) figures are averages of EA country R2 statistics weighted by
stock market capitalisation.

45
Chart 13: Shareholder rights
1998 2005

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: R. La Porta, F. Lopez-de-Silanes, A. Schleifer and R. W. Vishny (1998), “Law and Finance”, Journal of
Political Economy, OECD Corporate Governance and Company Law Database and ECB calculations.
Notes: The index ranges from 0 to 6, the lower the score the weaker are shareholder rights. The index is computed as
the sum of the following variables: (1) proxy by mail allowed; (2) shares not blocked before meeting; (3) cumulative
voting or proportional representation; (4) oppressed minorities mechanism; (5) pre-emptive rights; and (6) percentage
of share capital to call an extraordinary shareholder meeting. Variables from (1) to (5) equal 1 if allowed and 0
otherwise, while (6) equals 1 when the minimum required percentage is less than 20%, and 0 otherwise. For US for
2005 the update is not available. Data for 1998 have been appended. Euro area (EA) figures are averages of EA
country data weighted by stock market capitalisation.

46
Chart 14: Enforcement of shareholder rights against self-dealing (anti self-dealing index)
1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Djankov et al., 2006b and ECB calculations.


Notes: The index ranges from 0 to 1. Higher bars indicate better shareholder protection. The index incorporates ex-
ante and ex-post private control of self-dealing transactions. Ex-ante control includes the following issues for a
transaction between a corporate insider and an outside buyer: 1) Must disinterested shareholders approve the
transaction? 2) Must the buyer disclose the nature of the transaction and a possible ownership of the buyer by the
corporate insider?, 3) Must the corporate insider disclose the transaction and its nature? and 4) Is an independent
review, e.g. by a financial expert, required?. Ex post control considers the following points: 1) Are transactions
disclosed in periodic filings such as annual reports?, 2) Can a 10% shareholder sue the corporate insider for damages
suffered as a result of the transaction?, 3) How easy is it to rescind the transaction, 4) How easy is it to hold the
corporate insider liable for civil damages?, 5) How easy is it to hold approving corporate bodies liable for civil
damages?, 6) How easy is it to access evidence about the transaction? Euro area (EA) figures are averages of EA
country data weighted by stock market capitalisation. Data are from May 2003.

47
Chart 15: Creditor rights
1998 2002

0
AT BE DE ES FI FR GR IE IT NL PT EA SE CH UK JP US

Data sources: Djankov et al. (2006a) and ECB calculations.


Notes: The index ranges from 0 to 4. The higher the score the higher the protection. A score of one is assigned when
each of the following rights of secured lenders are defined in laws and regulations: First, there are restrictions, such as
creditor consent or minimum dividends, for a debtor to file for reorganisation. Second, secured creditors are able to
seize their collateral after the reorganisation petition is approved, i.e. there is no "automatic stay" or "asset freeze."
Third, secured creditors are paid first out of the proceeds of liquidating a bankrupt firm, as opposed to other creditors
such as government or workers. Finally, if management does not retain administration of its property pending the
resolution of the reorganisation. Euro area (EA) figures are averages of EA country data weighted by GDP.
Chart 16: Ownership concentration (size of shareholdings)
(fraction of shares held by the largest and by the first 10 shareholders)
Top 1shareholder Top 10 shareholders

70%

60%

50%

40%

30%

20%

10%

0%
AT BE DE ES IE FI FR GR IT LU NL PT EA CH SE UK JP US

Data sources: ECB calculations using Reuters Kobra database (2005).


Notes: Calculated on the basis of data available for the largest shareholders in 10 top quoted companies in terms of
market capitalisation in each country. The comparison highlights the concentration of share capital in the hands of the
first compared to the largest 10 shareholders. Euro area (EA) figures are averages of EA country data weighted by
market capitalisation.

48
Chart 17: Ownership concentration (Herfindahl index)
(Herfindahl index for the largest 20 shareholders)
30%

25%

20%

15%

10%

5%

0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: ECB calculations using Reuters Kobra database (2005).


Notes: Calculation of the Herfindahl index for the 20 largest shareholders in 10 top quoted companies in terms of
market capitalisation in each country. Euro area (EA) figures are averages of EA country data weighted by market
capitalisation.
Chart 18: Proportion of institutions among largest shareholder
(proportion of companies whose largest shareholder is an institution and proportion of institutions among ten largest
shareholders)
top 1 shareholder top 10 shareholders
1.0

0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0.0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: ECB calculations using Reuters Kobra database (2005).


Notes: Calculated on the basis of data available for the largest shareholders of the 10 top quoted companies in terms of
market capitalisation in each country. Institutional holdings are defined as holdings by buy-side institutions (the
investing institutions such as mutual funds, pension funds, and insurance firms). An institution is an entity in the
business of investment management (e.g., they employ investment professionals, have assets under management etc).
Investments may be managed on behalf of third parties or proprietary. Euro area (EA) figures are averages of EA
country data weighted by market capitalisation.

49
Chart 19: Law and Order index

1990-1994 1995-1999 2000-2005

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: International Country Risk Guide and ECB calculations.


Notes: The index ranges from 0 to 6, with lower scores indicating lower rates of law and order. The law sub-
component is an assessment of the strength and impartiality of the legal system, while the order sub-component is an
assessment of popular observance of the law. The index is the outcome of a subjective PRS staff analysis based on
questions such as: Are judges/magistrates appointed by qualification or by political affiliation/interest? How well paid
are police and law enforcement officers relative to other professionals? Have higher courts ruled against government
or against highly placed politicians/members of social/business elites? Euro area (EA) figures are averages of EA
country data weighted by GDP.
Chart 20: Duration of enforcement
(number of calendar days)
2002 2004

1400

1200

1000

800

600

400

200

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Djankov et al. (2003), World Bank and ECB calculations.
Notes: Total number of calendar days needed to recoup a bounced cheque, i.e. the time between the moment of
issuance of judgement and the moment the creditor obtains payment of a cheque. This is the sum of: (1) duration until
completion of service of process; (2) duration of trial; and (3) duration of enforcement. 2004 data are not exactly
comparable with 2002 data. The 2002 (2004) survey refers to a cheque worth the equivalent in local currency of 5%
(200%) of GNP per capita of the respondent country. The 2004 survey also considers administrative procedures for the
collection of overdue debt. Euro area (EA) figures are averages of EA country data weighted by GDP.

50
Chart 21: Enforcement of property rights
1995-2000 2001-2006

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: The Heritage Foundation and ECB calculations


Notes: the index ranges from 1 to 5. The higher the score, the lower the degree of property protection. The scores are
attributed according to the following criteria: 1= expropriation of property highly unlikely; 2= high, expropriation
unlikely; 3=moderate, expropriation possible but rare; 4=low, expropriation possible; 5=very low, expropriation
frequent. The criteria assessed are: State guarantee of property; efficiency of Court in contract enforcement;
independence of the judiciary system, existence of corruption. Euro area (EA) figures are averages of EA country data
weighted by GDP.
Table 2: Laws on Insider Trading
Country IT Laws Existence IT Laws Enforcement
AT 1993 n.a.
BE 1990 1994
DE 1994 1995
ES 1994 1998
FI 1989 1993
FR 1967 1975
GR 1988 1996
IE 1990 n.a.
IT 1991 1996
LU 1991 2004
NL 1989 1994
PT 1986 2003
CH 1988 1995
SE 1971 1990
UK 1980 1981
JP 1988 1990
US 1934 1961
Source: Bhattacharya, U., and H. Daouk, 2002, “The World Price of Insider Trading”, The Journal of Finance, 57,
pp. 75-108, and authors’ survey.
Notes: The numbers in the two columns came from the answers given to two questions sent to all the national
regulators and officials of stock markets of the world in March 1999. The two questions were: (1) When, if at all,
were insider trading (IT) laws established in your exchange? (2) If answer to (1) above is YES, when, if at all, was
the first prosecution under these laws? Since on the basis of the information available up to 1999 AT, IE, LU and PT
did not signal any prosecution of insider trading, competent authorities have been contacted by the authors in 2006
and asked to reply to question 2. No information available yet for AT and IE (n.a.).

51
Chart 22: Bank supervisor power
1999 2002 2005
14

13

12

11

10

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Information provided by supervisory authorities to the World Bank for 1999 and 2002, to the authors for
2005 and ECB calculations.
Notes: The index is on a scale from 0 to 14 with higher numbers indicating more supervisory power. The following
questions are quantified according to the answers yes=1 and no=0 and then summed up to yield the index: 1) Can
supervisors meet external auditors to discuss the report without bank approval?; 2) Are auditors legally required to
report misconduct by managers/directors to supervisory agency?; 3) Can legal action against external auditors be taken
by supervisor for negligence?; 4) Can supervisors force banks to change the internal organizational structure?; 5) Are
off-balance sheet items disclosed to supervisors?; 6) Can the supervisory agency order directors/management to
constitute provisions to cover actual/potential losses?; 7) Can the supervisory agency supersede bank shareholder
rights and declare bank insolvent?; 9) Does banking law allow the supervisory agency to suspend some or all
ownership rights of a problem bank?
Yes=1 and no=0 for the each of the following: Can the supervisory agency suspend director's decision to distribute: a)
dividends? b) Bonuses? c) Management fees?; Regarding bank restructuring & reorganization, can supervisory agency
or any other govt. agency do the following: a) supersede shareholder rights b) remove and replace management c)
remove and replace directors. No 2005 updates for FR,IT,NL,PT. Euro area (EA) figures are averages of EA country
data weighted by total assets of the banking sector. Due to data availability 2005 information had to be weighted with
2004 assets.

52
Chart 23: Supervisory forbearance discretion
1999 2002 2005
4

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Information provided by supervisory authorities to the World Bank for 1999 and 2002, to the authors for
2005 and ECB calculations.
Notes: The index ranges from 0 to 4. The higher the index the higher the discretion. The index is the sum of the values
given for the answers of the following questions. Does the law establish pre-determined levels of solvency
deterioration which forces automatic actions such as intervention? No =1. Can a supervisory agency or any other
government agency forbear certain prudential regulations? Yes=1. Must infraction of any prudential regulation found
by a supervisor be reported? No=1. Any mandatory actions in these cases? No = 1. No 2005 updates are available for
FR, IT, NL and PT. For BE and FR 2001 data are not available. Euro area (EA) figures are averages of EA country
data weighted by total assets of the banking sector. Due to data availability 2005 information had to be weighted with
2004 assets.

53
Chart 24: Supervisor independence
1999 2002 2005
4

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Information provided by supervisory authorities to the World Bank for 1999 and 2002, to the authors for
2005 and ECB calculations.
Notes: the index ranges from 0 to 4. The higher the index the higher the independence. Values for independence vs. the
political sphere are assigned according to the interpretation to the answers to the following questions (as 1=low,
2=medium, 3=high independence); 1) To whom are supervisors accountable?; 2) How is head of supervisory
agency/other directors appointed?; 3) How is head of supervisory agency/other directors removed?; The index is the
average of the answers to the questions above plus an assessment on the independence of supervisors vs. banks, i.e. the
protection against lawsuits from banks (yes=1). No 2005 updates for FR, IE, IT, NL, PT. Euro area (EA) figures are
averages of EA country data weighted by total assets of the banking sector. Due to data availability 2005 information
had to be weighted with 2004 assets.

54
Chart 25: Moral hazard index of deposit insurance
8

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK US JP

Data sources: Demirguc-Kunt and Detragiache (2002) and ECB calculations.


Notes: The index ranges from 0 to 8. The higher the index the higher the risk of moral hazard. Index values are the
sums of the values assigned to the following questions: 1) coinsurance required; 2) whether foreign currency deposits
are covered; 3) whether interbank deposits are covered; 4) whether deposit insurance is funded or not; 5) source of
funding (the scores are: 2 if government, 1 if government and banks, 0 if banks only); 6) type of management of
deposit insurance (the scores are: 3 if private, 1 if official, 2 if joint); 7) type of membership (the scores are: 1 if
compulsory, 0 if voluntary). The values assigned to points 1), 2), 3) and 4) are 1 if yes and 0 if no. No data are
available for ES and LU. Data refer mainly to the period 1999-2000. Euro area (EA) figures are averages of EA
country data weighted by total deposits.
Chart 26: Bank activity restrictions
1999 2002

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Information provided by supervisory authorities to the World Bank for 1999 and 2002, to the authors for
2005 and ECB calculations.
Notes: The index ranges from 0 to 4 and it is the sum of the scores in two categories, insurance and securities. The range
in each category is as follows: 0 = unrestricted, the full range of activities can be conducted or/but some or all must be
conducted in subsidiaries; 1 = restricted, less than full range of activities can be conducted in the bank or subsidiaries; 2
= prohibited, the activity cannot be conducted in either the bank or subsidiaries. Euro area (EA) figures are averages of
EA country data weighted by total assets of the banking sector.

55
Chart 27: Bank regulations supporting market discipline
1999 2002 2005
8

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Information provided by supervisory authorities to the World Bank for 1999 and 2002, to the authors for
2005 and ECB calculations.
Notes: The index ranges from 0 to 8 and the higher the score the better market discipline is supported by regulations. The
index combines information on the following categories: (1) external certified auditor required (yes=1, no=0); (2) No
explicit deposit insurance scheme (yes=1, no=0); (3) comprehensiveness of bank accounting (sum of the values (yes=1,
no=0) assigned to: (a) income statement containing accrued but unpaid interest/principal while loan is non-performing;
(b) consolidated accounts covering bank and any non-bank financial subsidiaries required; and (c) directors legally liable
for erroneous/misleading information; (4) off-balance-sheet items disclosed to public (yes=1, no=0); (5) banks disclose
risk management procedures to the public (yes=1, no=0); and (6) subordinated debt allowable (required) as part of
capital (yes=1, no=0). No 2005 updates for FR, IT, NL, PT, CH. Euro area (EA) figures are averages of EA country data
weighted by total assets of the banking sector. Due to data availability 2005 information had to be weighted with 2004
assets.

56
Chart 28: Insurance supervisory power

10

0
JP UK AT DE FR NL LU SE

Data sources: IMF FSAP and ECB calculations.


Notes: the index ranges from 0 to 10. The higher the index the higher the supervisory power. The index is based on the
compliance with applicable core principles of the International Association of Insurance Supervisors (IAIS). Values
assigned to the IAIS core principles are 1 if compliant and 0 otherwise. Compliance with the following core principles
is considered: ICP 11 deals with market analysis (authority monitors and analyses all factors that may have an impact
on insurers and insurance markets and takes action as appropriate); ICP 13 deals with on-site inspection (authority
carries out on-site inspections to examine the business of an insurer and its compliance with legislation and
supervisory requirements); ICP 14 deals with preventive and corrective measures (authority takes preventive and
corrective measures that are timely, suitable and necessary); ICP 15 deals with enforcement or sanctions (authority
enforces corrective action and, where needed, imposes sanctions based on clear and objective criteria that are publicly
disclosed); ICP 20 deals with liabilities (e.g. requirements for establishing adequate technical provisions and power of
authority to increase provisions if necessary); ICP 21 deals with investments (requirements of standards for investment
activities and enforcement of these by authority); ICP 22 deals with derivatives and similar commitments (authority
requires insurers to comply with standards on the use of derivatives); ICP 23 deals with capital adequacy and solvency
(authority requires insurers to comply with the prescribed solvency regime. This regime includes capital adequacy
requirements and requires suitable forms of capital that enable the insurer to absorb significant unforeseen losses); ICP
24 deals with intermediaries (authority sets requirements for the conduct of intermediaries); ICP 27 deals with fraud
(authority requires that insurers and intermediaries take the necessary measures to prevent, detect and remedy
insurance fraud).

57
Chart 29: Transparency of insurance sector regulation
7

0
JP UK AT DE FR NL LU SE

Data sources: IMF FSAP and ECB calculations.


Notes: the index ranges from 0 to 7. The higher the index the higher the transparency. The index is based on the
compliance with applicable core principles of the International Association of Insurance Supervisors (IAIS). Values
assigned to the IAIS core principles are 1 if compliant and 0 otherwise. The following principles are considered; ICP 2
deals with supervisory objectives (principal objectives should be clearly defined); ICP 4 deals with supervisory
process (authority should conduct its functions transparent and accountable; ICP 5 deals with supervisory cooperation
and information sharing (supervisory authority cooperates with other relevant supervisors); ICP 6 deals with licensing
(an insurer must be licensed before it can operate; licensing must be clear, objective and public); ICP 11 deals with
market analysis (authority monitors and analyses all factors that impact on insurers and insurance markets and takes
action as appropriate); ICP 25 deals with consumer protection (authority sets minimum requirements for insurers in
dealing with consumers in its jurisdiction, e.g. provision of timely, complete and relevant information); ICP 26 deals
with information disclosure and transparency towards the market (authority requires insurers to disclose information
on a timely basis in order to give stakeholders a clear view of their business activities and financial position and to
facilitate the understanding of the risks to which they are exposed).

58
Chart 30: Bank concentration
(Herfindahl index computed on total assets)
1998 2004

6000

5000

4000

3000

2000

1000

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Bankscope and ECB calculations.


Notes: The Herfindahl index has been computed by summing the squares of the market share of each bank in terms of
total assets. The index has been rescaled in order to range from 0 to 10000, with higher scores indicating more
concentrated markets. Euro area (EA) figures are averages of EA country data weighted by total assets of the banking
sector.

Chart 31: Foreign bank penetration


(Assets of foreign owned banks over total domestic assets, in %)
1999 2004

100%
90%
80%
70%
60%
50%
40%
30%
20%
10%
0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Worldbank for 1999 and ECB for 2004.


Notes: For 1999 data are missing for BE, FR, IE, NL and UK. For LU, CH, JP and US data are available only until
2002 from the Worldbank. Euro area (EA) figures are averages of EA country data weighted by total assets of the
banking sector.

59
Chart 32: Net interest income over total assets
1990-1994 1995-1999 2000-2003

4.0%

3.5%

3.0%

2.5%

2.0%

1.5%

1.0%

0.5%

0.0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: OECD and ECB calculations.


Note: Euro area (EA) figures are averages of EA country data weighted by total assets of the banking sector.

Chart 33: Net interest margins


(for commercial banks)
1996-1999 2000-2003

0.05

0.04

0.03

0.02

0.01

0.00
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: World Bank database on financial structure and ECB calculations.
Notes: Net interest margin is defined as difference between interest income and interest expense divided by interest
bearing assets. Country figures are unweighted country averages. Euro area (EA) figures are averages of EA country
data weighted by total assets of the banking sector.

60
Chart 34: Net non-interest income over assets
1990-1994 1995-1999 2000-2003

3.0%

2.5%

2.0%

1.5%

1.0%

0.5%

0.0%

-0.5%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: OECD and ECB calculations.


Note: Euro area (EA) figures are averages of EA country data weighted by total assets of the banking sector.

Chart 35: Operating expense over assets


1990-1994 1995-1999 2000-2003

5.0%
4.5%
4.0%
3.5%
3.0%
2.5%
2.0%
1.5%
1.0%
0.5%
0.0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: OECD and ECB calculations.


Note: Operating expenses are the sum of (i) staff costs, (ii) property costs and (iii) other operating expenses. Euro area
(EA) figures are averages of EA country data weighted by total assets of the banking sector.

61
Chart 36: Cost to income ratio
1990-1994 1995-1999 2000-2003

140%

120%

100%

80%

60%

40%

20%

0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: OECD and ECB calculations.


Note: Cost to income ratio is defined as operating expenses (sum of (i) staff costs, (ii) property costs and (iii) other
operating expenses) divided by the sum of non net interest income and net interest income. Euro area (EA) figures are
averages of EA country data weighted by total assets of the banking sector.

Chart 37: Return on assets


(net income after taxes over year end total assets)
1990-1994 1995-1999 2000-2003

1.5%

1.0%

0.5%

0.0%

-0.5%

-1.0%

-1.5%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: OECD and ECB calculations.


Notes: Euro area (EA) figures are averages of EA country data weighted by total assets of the banking sector.

62
Chart 38: Bank competition (H-statistic)
All Banks Commercial Banks

1.00

0.90

0.80

0.70

0.60

0.50

0.40

0.30

0.20

0.10

0.00
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: Bankscope and ECB calculations.


Note: the H statistic measures the elasticity of firms output to input prices. Under competition the H statistic is one,
under monopoly is equal or lower than 0. See section 3.7 for explanation on its computation. Euro Area (EA) figures
has been estimated by considering EA as a single country (i.e. using all banks from EA)

Chart 39: State ownership of banks


(percentages of total banking assets)
1999 2002

50%
45%
40%
35%
30%
25%
20%
15%
10%
5%
0%
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: World Bank survey conducted with national supervisory authorities and ECB calculations.
Notes: There are no data for IE. For BE, FR and SE data are missing for 2001. Euro area (EA) figures are averages of
EA country data weighted by total assets of the banking sector.

63
Chart 40: Economic freedom
1995 2000 2006

0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK EA JP US

Data sources: The Heritage Foundation and ECB calculations.


Notes: the index range from 0 to 5. The higher the score, the greater the level of government interference in the
economy and the less economic freedom a country enjoys. Euro area (EA) figures are averages of EA country data
weighted by GDP.

Chart 41: Control of corruption


1996 2000 2004

3.0

2.5

2.0

1.5

1.0

0.5

0.0
AT BE DE ES FI FR GR IE IT LU NL PT EA CH SE UK JP US

Data sources: World Bank database on governance and ECB calculations..


Notes: Higher values indicate better control of corruption. Average of indicators of corruption coming from a large
number of different sources (see Kaufmann et al., 2005, for details). It measures the exercise of public power for
private gain (e.g. nepotism, state capture or corruption). Euro area (EA) figures are averages of EA country data
weighted by GDP.

64