The French Co-operative Banking Group Model:

Too Good to be True?

Jean-Noël ORY 1
Yasmina LEMZERI 2

Corresponding author:
Jean-Noël ORY
Centre Européen Universitaire
15 Place Carnot
54 000 Nancy
FRANCE
[email protected]

Summary
This paper analyzes how French cooperative banking groups adapted their organization, status
and business model to develop and grow, up to the current financial crisis. It explores how
they benefitted from evolutions in cooperative law that lowered financing constraints and
increased the scope of their activities, thus becoming large banking groups, and identifies how
these groups tried to develop a model of governance, characterized by internal control, which
was partly dedicated to the members, but biased more and more towards the top of the
organizational pyramid and to stockholders (the new stakeholders coming from the existence
of listed vehicles). We also highlight how cooperative banks resisted potential threats in
proposed new international regulatory and accounting rules, often through a lobbying
network. While the developing business model for cooperative banks appeared to confer a
comparative advantage and was synonymous with efficiency before the financial crisis, it
seems now that the hybridization of the cooperative model has been a source of conflicts of
interest, of a weakness in strategy and an incentive to increase risk. This paper aims to
examine how and why these comparative advantages have become a burden, and whether all
French cooperative banking groups have suffered from the crisis in a similar way, or whether
different organizational and strategic features or choices may explain different levels of
resilience to financial turmoil.

Key words:
financial crisis, cooperative banks, governance, strategic choices, efficiency

1
Senior Lecturer (PhD) in Management, CEREFIGE, Université de Nancy 2, France.
2
PHD Student in Management, CEREFIGE, Université de Nancy 2, France.
 1

The French Co-operative Banking Group Model:

Too Good to be True?
Jean-Noël ORY 3
Yasmina LEMZERI 4

Résumé
Ce papier analyse la manière dont les banques coopératives françaises ont modifié leur statut,
leur structure organisationnelle et leur modèle, pour se developer, jusqu’à la crise financière
récente. Il relate notamment comment elles ont tire partie des evolutions en droit cooperative
pour diminuer leur contrainte de financement, et étendre et diversifier leur champ d’activité,
devenant ainsi de de grands groupes bancaires à part entière. Il explique également comment
ces groupes ont mis en place un mode de gouvernance spécifique, caractérisé par un contrôle
interne théoriquement dévolu aux sociétaires, mais de plus en plus dépendant du sommet de la
pyramide organisationnelle et des actionnaires, désormais présents au sein de ces groupes.
Nous mettons également en évidence que les banques coopératives ont jusqu’alors su résister
à des menaces potentielles provenant des nouvelles règles comptables et prudentielles
internationals. Alors que ces évolutions et choix semblaient leur conférer un avantage
comparatif et être synonyme d’efficacité et de performance, il semble aujourd’hui que
l’hybridation du modèle coopératif est également à l’origine de conflits d’intérêt, de choix
stratégiques discutables et d’incitation à la prise de risque. L’article tente donc également de
montrer dans quelle mesure les groupes bancaires coopératifs ont été affectés par la crise
financière, et si à des schémas organisationnels et des choix stratégiques différents répondent
un degré de résistance différent.

Mots-clés :
Crise financière, banques coopératives, gouvernance, stratégie, performance

3
Senior Lecturer (PhD) in Management, CEREFIGE, Université de Nancy 2, France.
4
PHD Student in Management, CEREFIGE, Université de Nancy 2, France.
 2

INTRODUCTION

Can cooperative societies compete efficiently with public limited companies (plc) in the
banking industry, and are they able to survive and to grow? These questions have been widely
discussed, particularly in the American and English literature of the late 1990s. Most studies
rely on the Agency Theory approach (JENSEN & MECKLING, 1976), and answer in the
negative, based on two main arguments: inappropriate status, and lack of governance.
According to them, cooperative banks are therefore doomed to disappear or to stagnate. In
Europe, and particularly in France, the situation seems to be quite different: cooperative
networks and groups have evolved differently and play an important role in the banking
industry.
After a brief description of the mainstream arguments against the efficiency of
cooperative banks, the second part of this paper aims to analyze how French cooperative
groups have adapted their organization, status and business model in order to survive in a
competitive environment. More precisely, how they’ve been able to benefit from changes in
cooperative law, lowering their financing constraints and increasing the scope of their
activity, thus becoming large banking groups. Simultaneously, these groups have tried to
develop a model of governance typically characterized by internal control that is partly
assigned to the members (but increasingly transferred to the top of the organizational
pyramid), and partly assigned to the stockholders (the new stakeholders resulting from the
creation of listed vehicles). They've also been able to adapt, and have resisted the threat of
some proposed changes in international regulatory and accounting rules, often through a
lobbying network.
Whereas this developing business model appeared to represent a comparative
advantage before the financial crisis and was synonymous with efficiency, it seems now that
the hybridization of the cooperative model has been a source of conflicts of interest, a
weakness in strategy and an incentive to increase risk. Therefore, the third part of this paper
aims to examine how and why these comparative advantages have become a burden, and
whether all French cooperative banking groups have suffered in a similar way during the
crisis, or whether different organizational and strategic features or choices may explain
different levels of resilience to the financial turmoil.
 3

1. THE THEORETICAL WEAKNESS OF THE COOPERATIVE MODEL IN THE

BANKING INDUSTRY

Several studies have shown that organizations can be characterized by different ownership
and control structures, and therefore by different governance mechanisms, which have an
impact on their efficiency. According to Agency Theory and Neo-Institutional Theory, the
efficiency of an organization depends on its ability to reduce transaction and agency costs
(JENSEN & MECKLING, 1976; WILLIAMSON, 1983). Only the most efficient
organizations are expected to survive in a competitive environment. In line with FAMA &
JENSEN (1983), the classical, traditional Anglo-American model of governance relies on
shareholder value, and considers that the most efficient organization is the one that grants
control to the “residual claimant”, that is to say, the shareholder. In such a firm, the
shareholder has a direct interest in maximising the global value created by the firm, because it
is the best way to maximise his or her own profit, i.e, the residual value of the firm, on which
they have a property right. The shareholder’s status of “owner” legitimates his or her control
over the manager, and therefore the requirement to check that decisions taken by the latter are
the best choices for the firm and for the shareholder. Consequently, the ability of corporate
governance to solve the classic agency conflicts (managers/shareholders,
shareholders/creditors…) depends on the existence of appropriate control mechanisms. For
this type of efficient governance, internal mechanisms are helpful (reporting, board of
directors with independent board members…), but external mechanisms are considered
essential (CHARREAUX, 1997a and 1997b): the disciplinary role assigned to the financial
market is particularly important, because it allows mechanisms of rewards and sanctions, and
signals the efficiency or otherwise of the firm, as long as the financial markets are sufficiently
deregulated.
This simplified version of the theoretical framework, although modified more recently
(JENSEN 2001), has been the foundation for the work of several theoreticians. Several years
ago, before the financial crisis, some of them even argued that this model of governance,
thanks to its efficiency, would spread all over the world (HANSMMAN & KRAAKMAN,
2001), and would replace alternative models existing in Japan and particularly in Europe (O’
SULLIVAN, 2002), where cooperative firms coexisted with joint stock companies. Most of
these theoreticians thought that cooperative banks suffered from a lack of efficiency, and this
explained their weak performance. Indeed, a commercial bank and a cooperative bank don’t
deliver the same property rights to their owners. In contrast to a shareholder in a plc, a
 4

member of a traditional cooperative owns a share that is not listed, whose value is not related
to the residual value of the firm, whose return is limited 5, and which is not easily traded. In
many countries, and particularly in France, cooperative members have no individual rightful
claim on the reserves, which belong to the cooperative itself. Generally, the rule “one person,
one vote” prevails, instead of “one share, one vote” 6. Moreover, a member of a cooperative
has a twin identity, being simultaneously an owner of the cooperative and a consumer (thus,
as a customer of the bank, the member is also a creditor). Therefore, motivation for members
is not only maximisation of individual wealth as an owner of the capital, but also possibly to
have a “return” as a consumer of the financial products and services 7. The common bond
shared with the other members is also assumed to be an incentive.
According to Agency Theory, the common bond associated with the twin identity of
the member should allow cooperative banks to benefit from a comparative advantage in
reducing information asymmetry (Figure 1). Thanks to the homogeneity of the customers and
the existence of a strong loan relationship, the individual credit risk in the creditor-debtor
relationship can be reduced in comparison with plc banks (HANSMANN 1985; HART &
MOORE, 1990; BERGER & MESTER, 1997). Nevertheless, it is impossible for a
cooperative bank to diversify the entire credit risk because of its small size, and its inability to
benefit from economies of scale or scope and the difficulty of raising capital would explain
the problems involved in surviving in a changing environment (AKELLA & GREENBAUM,
1988; MESTER, 1993). Moreover, these comparative advantages are expected to disappear as
a cooperative bank grows.

5
The dividend of the share is quite similar to an interest rate, with a statutory or legal cap.
6
This rule prevails at least at the local level of the cooperative bank, although sometimes partly modified at the regional or
national level.
7
If we consider the emergence of the cooperatives in Europe, especially in banking, one of the explanations has to be found
in the need for some social groups or individuals to accede to financing. One of the original principles in such organizations
for the member was to benefit from refunds, rebates, whose importance were linked to his use of the services offered by the
cooperative.
 5

Figure 1
The common bond of the customers: advantages and constraints

Common bond
Homogeneity of the members/customers

Decrease in the asymmetry Constraint: Constraint:

of information simple and standardized small size of the coop.
banking activity banks
Cohesion in the goals of the
organization

antagonism antagonism

Changes in the banking environment

globalization of the markets, decreasing intermediation,
increasing need in sophisticated financial products
allows

prevents

Efficiency and growth of coop. banks

The same American authors also argue that the internal and external mechanisms of
governance are actually inefficient (Figure 2): members are not encouraged to control the
managers because of diffuse ownership rights, and because of a weak correlation between the
net profit of the cooperative and the return on member shares. These features could increase
the managers’ discretionary powers and the firm’s inefficiency (MAYERS & SMITH, 1994).
The twin identity also creates a confused situation, particularly in defining the goals of the
firm. Financial market discipline is unlikely to play a role because the cooperative bank is not
listed, hence no reward/sanction due to changes in share price, and no threat of takeover bids.
Consequently, personnel costs are likely to stand at a high level and productivity is expected
to be lower than in conventional plc banks (AKELLA & GREENBAUM, 1988); more
generally, cost-cutting incentives and financial efficiency (return on equity) could be
insufficient.
 6

Figure 2: The inefficiency of governance mechanisms in cooperative banks

(Agency Theory approach)

Low return of the member- Member

Limited ownership shares and weak correlation with shares not
rights of the members the result of the coop. bank listed

- No financial market
No incentive for the members
Weak correlation discipline
to control
between the
composition of the - No risk of takeover
board of directors bids
and the ownership
of the shares Weakness of the control
on the managers
No incentive in
Discretionary the reduction
managerial Waste
Opportunism in Overstaffing of costs and
decision-making expenses
power investment
choices Low productivity

Organizational inefficiency
Impossibility to compete efficiently
Weakness of the efficiency

The situation of cooperatives in the banking industry in Europe, and particularly in France,
goes against the arguments mentioned above. First, if we consider the size of the cooperative
networks, we can see that these banks own important market shares in deposits and credit
financing (FONTEYNE, 2007). Secondly, unlike credit unions in the USA, many of the
European cooperative banks are structured in networks with different levels (local/national;
local/regional/national), as is the situation in France. This structure enables these cooperative
networks to combine the advantages of small banks with those of large networks.

2. THE HYBRIDIZATION OF THE COOPERATIVE MODEL AND THE

EFFICIENCY OF FRENCH COOPERATIVE BANKING GROUPS

In this section, we try to explain how French cooperative banks developed and responded to
the assumed constraints listed above. Having grown and evolved into large banking groups,
the cooperative model has changed and has become hybrid. We end the section by showing
that, before the 2007 crisis, the different French cooperative banking networks and groups had
 7

comparable efficiency to the plc banks, even though there were differences in organizational
schemes and levels of hybridization among the different cooperative groups.

2.1. A GROWING RANGE OF FINANCIAL INSTRUMENTS LOWERING DEVELOPMENT

CONSTRAINTS

As mentioned, many of the theoretical Anglo-American studies that analysed cooperative

status in banking pointed out that difficulties in raising capital limited the possibility of
development. However, French cooperative banks have benefitted from changes in
cooperative law that have lowered this constraint, without allowing internal and external
control to change hands. A law enacted in 1992 allows non-members to own part of a
cooperative’s capital 8, and even to subscribe to specific shares with higher returns, if these
partners relinquish their right to vote. It also gives a cooperative the opportunity to re-evaluate
the face value of shares, to issue specific member shares with higher returns, or to distribute
bonus shares to members by incorporation of reserves, in order to compensate for the weaker
rights of members to the net assets, compared to shareholders in a plc company.
Several other legal changes have also made it possible for cooperative firms to issue
hybrid financial instruments, such as participating securities (1983), member cooperative
certificates (1992) and investment cooperative certificates (1987). These financial products
are characterized by a return that is more correlated with the benefit, via a market value
evolving on a secondary market, but do not bring any right to vote 9 to their holders.
In the French banking industry, only a few of the growing range of financial instruments
available have been used significantly by cooperative banks. French cooperative banks appear
to have been reluctant to expand the issue of specific shares, increasing returns or extending
rights of members (HIEZ, 2005), especially on the reserves. Apart from some marginal
experiments lead by some of the Banques Populaires, none of the French cooperatives have
chosen to re-evaluate the face value of their shares, and where some cooperative networks
offer specific shares to their members (Credit Mutuel, Credit Coopératif), these generally do
not bring a higher return from incorporating reserves, but are cut off from the right to vote, or
coexist with member shares that bring no return at all. Participating securities don’t exist, and,
except in the case of several regional Caisses de Credit Agricole, member or investment

8
Normally, this part is caped to 35% of the rights to vote; the limit is possibly increased to 49% when the partners are also
cooperative societies.
9
The whole of these hybrid instruments, added to the specific shares, cannot exceed 50% of the capital of the cooperative.
 8

cooperative certificates are not offered to the members/customers. In fact, it seems that French
cooperative banks have been conscious of the strategic importance of reserves to their internal
and external growth as firms and for compliance with Basel capital requirements; therefore,
any instrument operating a cut-off on the reserves to the benefit of members or other
stakeholders has been considered as a potential threat, rather than as a useful tool for
development. Alongside this is their desire to maintain internal control of the cooperative
firm, and avoid a spread of rights among members or other stakeholders.
The possibility for non-members to own part of the capital has been more widely used by
French cooperative banks because it allows the capital funding constraint to be lightened,
without sharing property rights. Furthermore, in line with plc banks, the central bodies of the
different French cooperative banking networks have also chosen to issue specific
subordinated debt instruments, whose characteristics allow them to be considered as
regulatory equity capital (tier 2 or 3), without control having to be shared.
Three of the four main French banking cooperative networks 10 have also chosen to issue
cooperative investment certificates. However, the logic behind the use of these hybrid
instruments is different to what was anticipated by the laws enacted in 1987 and 1992: they
have been issued by regional cooperative banks, and have not been subscribed to by the
customers, but by the apex (CASA, in the case of Credit Agricole) or by a subsidiary holding
(Natixis, in the case of Caisse d’Epargne and Banque Populaire). The objective is thus not to
let members/customers participate in the results of the regional banks, but to allow part of the
value created by the regional banks to be consolidated by the apex or holding subsidiary at an
aggregated level, and to be distributed to other stakeholders (especially stockholders of the
cooperative groups). Cooperative investment certificates have thus been used as a
development tool embedded in a larger group development strategy.

2.2 THE BUILDING OF LARGE COOPERATIVE BANKING GROUPS

The changes in organisational charts and in strategic and business models are a key feature in
explaining the increased efficiency of the French cooperative banks and their growing role in
the banking industry. Although some of the characteristics mentioned below can be found in
other European countries (OP group in Finland, or Raiffeisen in Austria), France is probably
the only country in Europe where all the banking cooperative networks have experienced such

10
This is the case for Caisse d’Epargne, Banque Populaire, Credit Agricole, but not for Credit Mutuel.
 9

deep changes. These networks have been transformed into large cooperative banking
groups 11, and we support the thesis that it is more and more difficult to distinguish the activity
of the cooperative banks themselves from that of the group. Unfortunately for some of them,
the financial crisis has revealed some possible problems associated with overlapping
activities, as we will discuss in Section 3.
The first key feature of the evolution of cooperative banks in France was the change in
legal form of the pyramidal cooperative network at the national level. For Credit Agricole,
Caisse d’Epargne and Banque Populaire, the central institution (Caisse Nationale or Banque
Fédérale) has become a conventional joint stock company, whereas the regional/local banks
have kept their cooperative status. Despite an organizational chart for Credit Mutuel that is
yet more diffuse, this is also the case for the Banque Fédérative du Credit Mutuel, which is
controlled by the main regional cooperative bank of this group.
The joint stock company legal form is more appropriate for restructuring arrangements
(such as buying other banks, via takeover bids or share exchanges, and putting these
subsidiaries together under centralized control), but the cooperative banks still keep the
majority interest in the central body, therefore avoiding the threat of takeover bids
themselves. Furthermore, while the existence of a listed vehicle is a common point for all of
the French cooperative banking groups, different strategic choices have been made. Whereas
the central institution is listed in the case of the Credit Agricole group (CASA), it is the
holding subsidiary in the Caisse d’Epargne/Banque Populaire group (Natixis, whose control is
equally shared by the central institutions BFPB and CNCE 12) that is listed, and CIC 13, the
only listed vehicle in the Credit Mutuel Group, is held by a subsidiary of a regional bank, and
not by the central institution. The choice of public listed vehicle has been another way to
lower financing constraints, thanks to the possibility of stock issues; and possibly also thanks
to providing better information signals to rating agencies (since a favourable rating can lower
the cost of debt).
The large amount of reserves, more easily accumulated thanks to the cooperative legal
form and the existence of a “solidarity principle” that enables the head of the group to raise
capital through the regional banks if necessary, have facilitated the creation and, mainly, the
acquisition of various subsidiaries at a time (in the 1990s) when conventional plc banks were

11
The « network » refers to the perimeter of activity of the local/regional cooperative bank themselves, whereas the « group »
includes the activity of the different subsidiaries whose ownership is directly or indirectly hold by the network.
12
CNCE: Caisse Nationale des Caisses d’Epargne, central institution of the Caisse d’Epargne Group.
BFPB : Banque Fédérale des Banques Populaires, central institution of the Banque Populaire Group.
13
CIC: Crédit Industriel et Commercial, a holding bank that owns different regional joint stock banks, and which is owned
itself by the Banque Fédérative de Crédit Mutuel, subsidiary of the main Caisse Regionale de Crédit Mutuel.
 10

experiencing financial problems. These subsidiaries can be divided into different sub-groups:
those that provide direct technical support to the group (typically IT platforms) and whose
objective is an increase in scale economies; “production subsidiaries”, which develop new
financial products and manage them for the whole retail banking network; specialized
subsidiaries, whose objective is to offer and manage sophisticated products and services
(private banking, financial engineering etc); subsidiaries located abroad; and other financial
institutions specializing in market finance and investment banking.
The presence of listed vehicles within the groups and the existence of subsidiaries whose
objective is to add value have also changed the business model of the cooperative groups.
While the local and regional bank administrators and members still defend the cooperative
principles and values, the objectives of “return on equity” and “maximisation of financial
profitability” have been introduced via a “management by objectives” policy and RAROC
(risk adjusted return on capital) methodology. The role of the central institution has increased,
and cost-cutting has been encouraged; this evolution also explains the level of efficiency,
according to conventional ratios and excluding social economy criteria.
Finally, the co-existence of cooperative banks and conventional plc banks in the economy,
of listed vehicles besides member-owned local banks, of retail and small and medium industry
(SMI) banking besides international financing and engineering, has led to the development of
a hybrid model and a “universal banking” strategy among large cooperative banking groups.
This hybridization conferred a comparative advantage prior to the financial crisis, for it
simultaneously allowed the cooperatives to benefit from both cooperative and plc bank
characteristics, and proved their adaptability in a changing banking environment.

2.3 THE ABILITY TO ADAPT AND TO RESIST POTENTIAL THREATS FROM THE REGULATORY

AND ACCOUNTING ENVIRONMENT

Despite hybridization, French cooperative banking groups have been able to resist potential
threats and defend their specificity whenever necessary. In the face of a growing trend
towards normalization affecting the regulatory and accounting environment in the banking
industry, the cooperatives have organized themselves into a lobbying group at the national
level, via the Groupement National de la Coopération (GNC), and even at the European level
via the European Association of Cooperative Banks (EACB). As a consequence, they have
 11

been able to influence the development of standardized rules and prevent such rules from
lowering their competitiveness.
The first example is to be found in the original Basel Committee proposals, from the
early 2000s. The new prudential rules, which relied on estimation of credit risk given by
rating agencies, were weighted in favour of the “big banks” whose activity was dedicated to
large corporate firms. The cooperative banks pushed the Basel Committee to consider an
alternative definition of the weighting coefficients used to calculate the minimum regulatory
capital requirement. More precisely, this new definition took into account the risk
diversification principle within the customer portfolio (characterizing the small but numerous
credits of the cooperative banks, whose activity is in retail or SMI banking); consequently, it
allowed a lower prudential equity requirement.
The second example concerns the new international accounting rules. In June 2002,
the International Accounting Standards Board (IASB) 14 published a new amendment
dedicated to financial instruments (IAS 32). According to this amendment, “each financial
product that enables the holder to be paid back at his request, has to be considered as a debt
instrument” (DETILLEUX & NAETT, 2005). Consequently, the member shares issued by the
cooperative banks could have been considered as debt and no longer as equity. Given the
importance of the threat, the cooperative banks, relying on EACB lobbying action, pushed the
IASB to modify the conditions defining equity capital in banking (DETILLEUX & NAETT,
op. cit.; EACB, 2006). This question is still under discussion, since the IASB and the
Financing Accounting Standards Board (FASB) wish the American and IFRS accounting
rules to converge (EACB, 2007), and also because of a possible change in the Basle
Committee definition of the equity-capital.
Thanks to their adaptability and their reactivity, French cooperative banks became able
to stand comparison with conventional plc banks, and, before the crisis, were characterized by
efficiency and a singular position in Europe.

2.4 THE COMPARATIVE EFFICIENCY OF THE COOPERATIVE NETWORKS AND GROUPS AND
THE PLC BANKS

The goal of this section is double: first, to compare the efficiency of the French cooperative
networks and groups between themselves and second, to compare them with the efficiency of

14
Part of the International Standard Accounting Standard Committee, in charge of the definition and appliance of the new
international accounting rules, also called IFRS (International Financing Reporting Statement)
 12

plc banks, over the period 1994-2004. We have deliberately taken efficiency indicators into
account that are commonly used to assess the performance of conventional banks (Bankscope
database), and not social efficiency indicators. The indicators can be ordered into five
categories: credit risk ratios, capital equity ratios, commercial performance ratios, operational
ratios and financial ratios. The contents of these indicators are detailed in Appendix 2. For
each of these ratios, Wilcoxon paired rank tests were conducted on each of the sub-samples
under consideration (i.e cooperative networks/groups or plc banking groups). Then, the results
were ordered by building a performance score (the larger the score, the greater the
performance, as explained in Appendix 3).

2.4.1 The aggregated analysis: French cooperative networks/groups versus plc

banks
Two subsamples of cooperative banks were built, the first representing aggregated
performance ratios of all the French banking networks, i.e, excluding the activity of the
subsidiaries, and the second including the activity of the subsidiaries within the group
(investment banking etc). These subsets were compared to two plc bank subsamples: the
first made up of French banks, and the second of the 30 most important European banks
by total assets (see Appendix 1 and Appendix 4).

A better global efficiency for French cooperative banks than French plc banks…
Over the last decade, the tests lead to the conclusion that the management of credit risk
has been at least as efficient in French cooperative banks as it has been in French plc
banks, and even better: the ratio loan loss reserve/impaired loans is not statistically
different between these two kinds of banks, but the loan loss provision/net interest
revenue ratio is in favour of the cooperatives. This result is confirmed by the fact that the
ratio net interest revenues/average assets is also significantly larger, which denotes that
the cooperatives were able to set credit rates appropriate to an effective credit risk.
In view of the operational efficiency ratios, we can say that, despite a higher personal
expenses/total assets ratio, which could confirm some of the arguments listed in Section
1, the cooperative banks generated higher operating incomes compared with the amount
of assets. This explains how the cost to income ratio is not statistically different between
the two categories of banks, thus, the argument of lower operational efficiency for
cooperative banks can be refuted.
 13

The equity capital ratios are clearly in favour of the cooperatives: the equity/total
assets ratio appears statistically higher. It seems that they do not suffer from a lack of
capital funds, as mentioned in the American analysis, and that cooperative status, thanks
to the high level of reserves and the existence of member shares, has to be seen as a
comparative advantage rather than a disadvantage in France.
Lastly, a more surprising conclusion is that, even if the return on equity ratio is in
favour of the French plc banks (and also much more volatile), the return on average
assets ratio appears larger for the cooperatives.

…The evolution towards universal cooperative banking groups does not improve
efficiency, but still allows some comparative advantages.
In light of the equity capital ratios, the consequences of transforming cooperative banks
into universal banking groups are clear: the cooperative groups benefit from a higher
capital ratio (equity/total assets) than plc banks, thanks to the magnitude of this ratio in
the cooperative network. However, credit risk management and commercial efficiency
ratios are lower, when taking the perimeter activity of the groups into consideration, and
can be considered as statistically equal to the French plc bank indicators. Considering
operational efficiency, the tests show that personnel expenses are relatively lower in the
cooperative groups than in the cooperative banks, but are still higher than in French plc
banking groups. Nevertheless, the cost to income ratio is not significantly different.
If we focus on financial efficiency indicators, we notice that, by the end of the decade
1994-2004, the return on equity criterion was still in favour of the French plc banks,
whereas the return on assets has decreased, comparing to that of the perimeter activity of
the cooperative banks. By this stage, the growth strategy of cooperative banks and the
move towards huge groups have not brought the same returns to the shareholders of these
groups, compared with those of French plc banks. The reasons behind the evolution of
cooperative banks towards universal banking groups therefore has to include factors other
than financial efficiency alone (economies of scale or scope, position in the competitive
environment…, see ORY, JAEGER & GURTNER, 2006b).

Are European plc banks more efficient than French banks?

The European sample of plc banks does not seem to be more efficient than the French
cooperative banks if we consider credit risk management ratios: the tests point to equality
 14

in the ratios observed. Commercial performance ratios tend to show that the effective risk
is properly hedged by the rates applied to customers, in a better way than in the French plc
banks or cooperative groups. Operational efficiency indicators show that pre-tax operating
income is larger for the European plc banks than for French plc or cooperative banking
groups. Only French cooperative banks equal this ratio. Even so, European plc banks are
characterized by a lower cost to income ratio than French cooperative networks, groups or
plc banks. While the tests show that French cooperative banks hold a comparative
advantage in equity capital ratio, this is not the case for French plc banks. Finally, a high
level of operational, commercial and risk-management efficiency explains the good
financial performance of the European bank sample over this period.

2.4.2. Comparing French cooperative networks and groups

Conducting the same Wilcoxon tests on the different French cooperative networks and
groups gives us more detailed results. First, we notice that the different French
cooperative networks (excluding the activity of subsidiaries) do not exhibit the same
characteristics (see Appendix 5). Whereas Caisse d’Epargne is characterized by the best
credit risk ratio over the period (mainly explained by retail banking activity), the Banque
Populaire and Credit Agricole networks are associated with the best operational,
commercial and financial efficiency, and the highest equity capital ratio.
Taking into account the perimeter activities of the group (including subsidiaries)
modifies the conclusions. The equity, commercial and operational efficiency ratios of
Credit Mutuel improves significantly, although the return on equity remains low. Whereas
commercial, operational and financial efficiency has substantially deteriorated for Credit
Agricole, Caisse d’Epargne is still characterized by a low credit risk and low operational
performance. Finally, regarding commercial, operational and financial criteria, Banque
Populaire can be still considered as the best performing group, although credit risk
management and equity capital ratios have fallen. These characteristics have changed
dramatically during the recent financial turmoil.
 15

3. IS THE HYBRIDIZATION OF THE COOPERATIVE MODEL SYNONYMOUS

WITH A LOWER RESILIENCE IN FACE OF THE FINANCIAL CRISIS?
The first part of the 1990s was already synonymous with a crisis period, in the French
banking industry, and especially for the plc banks. At that time, French cooperative banks
were globally characterized by higher equity ratios, and above all, by a better risk-efficiency
(Appendix 8). This resilience offered them the opportunity to build some of these plc banks
and to include them into a larger perimeter of activity (thus to transform themselves into
cooperative groups – ORY, GURTNER & JAEGER, 2006b). This is not the case anymore.
Actually, while the French situation can be considered as quite singular in the European
banking industry due to the importance of cooperative groups and their successful
development model, it has not been immune from the recent financial turmoil. This third
section aims to analyze how the crisis has affected the French cooperative groups, comparing
them to each other and also to two other French plc groups. We then try to identify the lessons
that can be drawn from the crisis regarding the hybridization of French cooperative banking
groups.

3.1 THE FACTS: THE FRENCH COOPERATIVE BANKING GROUPS AND THE FINANCIAL CRISIS

Since the current financial turmoil only started in summer 2007, several of the relevant data
are not yet available on the Bankscope database. In this case, we have usedthe indicators that
published by the groups themselves in their annual reports. Moreover, the cost of risk is still
evolving in the banking groups, for it is still difficult for them to assess the cost and the
amount of the “toxic assets” : consequently, the reader should know that the confidence level
of the data in 2007-08 is not as high as it is before the crisis . The available data
include:banking net product (BNP), gross operational result (GOR), net result (NR),
operational result (OR), cost of risk (CR), cost to income coefficient (CI), and return on
equity (ROE) (see Appendix 7). In addition to these, we have taken into account different
ratios in order to make comparison easier (see Appendix 8). We have considered not only the
data for each of the banking groups, but also for their different activities, i.e. retail banking on
one hand and corporate and investment banking (CIB) on the other. This enables us to check
whether all activities have been equally affected by the crisis.

3.1.1 The financial crisis has affected all bank results in terms of value…
 16

The first conclusion we can draw is that the results of all the French banking groups have
been affected in value since 2007, whether cooperative or plc. The only one that has been
relatively untouched is BNP-Paribas. Among the different activities in these groups, corporate
and investment bank activity (CIB) is of course the most affected, even among cooperative
groups. Considering CIB activity alone, two groups are characterized by negative results in
2007: one is a plc group, but the other is a cooperative group, being Societe Generale, net
result (-2221 millions Euros 15), and Credit Agricole 16, operational gross result (-756 million
Euros), operational result (-1713 million Euros) and net result (-904 million Euros).
In 2008, the situation worsened, as all the French banking groups investigated suffered
from negative CIB results, and there were even losses at the group level for some. Indeed, the
operational (but not net) result for CIB for BNP-Paribas became negative for the first time (-
1215 million Euros). The cooperative groups have not been immune. The gross, operational
and net results for Credit Agricole were negative for CIB activity, and so were the same
indicators for CIB activity in Credit Mutuel 17. Due to the huge losses experienced by their
common subsidiary Natixis, Caisse d’Epargne and Banque Populaire saw their own
operational and net results, as well as their return on equity ratio, becoming negative at the
group level, and not just CIB perimeter activity.
There is strong evidence that retail banking activity has contributed to the resilience of
the cooperative and plc groups. Moreover, when the data are detailed, it is also obvious that
retail banking activity operated by plc banks within a cooperative group (for example LCL in
Credit Agricole or the regional banks held by CIC in Credit Mutuel) has provided similar
resilience. Expressed in ratios (see below), this lesson is confirmed, but it is noticeable that
even retail banking activity has been affected by the crisis.

3.1.2 The incidence of the crisis expressed in efficiency ratios

As we’ve seen, corporate and investment banking activity is the principle source of the fall in
results for cooperative and plc groups. However, when we consider each growth rate indicator
separately, and for each activity (whole group, retail and investment bank), all indicate a drop
in results. But, expressed in value, these results are difficult to compare, so we have built
efficiency ratios to provide more in-depth analysis.

15
The losses coming from the Jerome Kerviel illegal activity are excluded from this result.
16
The CIB activity of Credit Agricole group comes mainly from the subsidiary CALYON.
17
A large part from Credit Mutuel corporate and banking activity comes from its plc subsidiary Credit Industriel et
Commercial.
 17

The cooperative groups have suffered a lot …

Based on our efficiency ratios, Credit Mutuel exhibited very good results in 2006 and 2007,
but not in 2008, whatever ratio is considered. Credit Mutuel’s operational gross result/bank
net product 18 fell over one year. These bad results are explained by the fact that they take CIB
activity into account, which was strongly affected by various factors. First of all, the group
holds a commercial bank subsidiary, CIC, which conducts CIB activity under the CM4-CIC
name. This subsidiary was affected by the Madoff Ponzi scheme, losing 110 million Euros.
Secondly, the majority of the loss, some 500 million Euros, is explained by the Lehmann
Brothers bankruptcy, to which the group was highly exposed via CM4-CIC. Lastly, the fall of
the Icelandic banks cost the group another 65 million Euros. Operational results/bank net
product 19 also suffered in 2008 from losses associated with the rise in the cost of risk to the
group, particularly around CIB activity.
Compared with Credit Mutuel, Credit Agricole did relatively well in 2008, despite the
losses incurred by its subsidiary Calyon, and especially if we consider the operational gross
result and the net result as a percentage of net product. But regarding operational result, which
has fallen due to the cost of risk, this is no longer the case, meaning that the cost of risk was
high, as confirmed in Appendix 8. The cost of risk for Credit Agricole grew dramatically
mainly because of the losses it suffered through one of its Greek subsidiaries, Emporiki. It
was also hurt directly by the subprime crisis via its CIB subsidiary, Calyon, which cost it 6
billion Euros, and losses due to the bankruptcy of Lehman Brothers and Madoff amounted to
200 million Euros.
As mentioned previously, the Banque Populaire and Caisse d’Epargne groups have in
common the fact that they are both owners of Natixis, a corporate and investment bank.
Banque Populaire and Caisse d’Epargne are not built on the same model, and it is interesting
to see whether or not they have suffered in the same way from the crisis. As regards the two
first ratios 20, Banque Populaire and Caisse d’Epargne exhibit the worst efficiency ratios,
mainly because of the results from Natixis. Indeed, Natixis has clearly been exposed to the
subprime crisis, and its monocline firm, CIFG, was also hurt directly. In 2008, its negative
contribution to the Caisse d’Epargne result amounted to about 2 billion Euros. It also lost 1.2
billion due to financial portfolio value adjustments. Finally, a 3 billion Euro loss was recorded

18
This ratio represents the bank net product minus the operational fees related to the bank net product
19
The operational result is the operational gross result reduced of the cost of risk.
20
Operational gross result/bank net product and operational result/bank net product.
 18

in 2008 due to investment activity at the regional bank network level. Because of Natixis,
Banque Populaire group lost 1.3 billion Euros in 2008.

… whereas the French plc banks have shown a relative better global resilience.
It seems that the plc bank groups, i.e. Societe Generale and BNP, have handled the crisis
better. Indeed, in 2008 and using our three ratios, BNP Paribas is the best group, followed by
Societe Generale. Despite the losses they recorded, they managed to maintain ratios because
they did not have the same bank net product (BNP) as the cooperative groups. As they
realized a greater BNP and as their total assets are bigger, even if they recorded a loss of the
same amount as their cooperative counterparts, the relative impact would be less important.
They also succeeded in improving all their solvency ratios and show the best International
Basel 2 ratio (11.1% for BNP and 11.6% for Societe Generale in 2008).

3.2 LESSONS FROM THE CRISIS REGARDING THE HYBRIDIZATION OF THE COOPERATIVE

MODEL

Besides the direct effects of the crisis on the results of cooperative groups, several lessons can
be drawn regarding the hybridization of the cooperative model in the banking industry.

3.2.1 The cooperative legal form is not sufficient protection against the crisis
The first lesson is that the cooperative legal form is not a sufficient protection against the
effects of the financial crisis in itself. It seems more appropriate to say that retail banking
activity and traditional corporate financing (via credit) are characterized by a lower risk and a
lower volatility compared to investment banking or to activity on the financial markets. Of
course, since retail banking and traditional financing to SMI are the heart of the business for
local/regional cooperative banks, these can be considered as being a significant stabilizing
component in the face of a financial shock, and have clearly contributed to mitigating the
consequences of the crisis on the results of cooperative groups. But the same activities have
played a similar role within the conventional plc banking groups (BNP and Société Générale),
even though the commercial activity of these networks does not rely on cooperative banks.
So, the contribution of cooperative banks to improving the resilience of the banking and
financial system also depends on their ability to be dominant, not only in the activity of the
group to which they belong, but also in the strategic and organizational choices made by the
group and in its governance model.
 19

In France, greater resilience to the crisis seems to be found in the Credit Cooperatif
group, the smallest (it has been part of the Banque Populaire Group since 2003, but is still
distinct in some key features), whose activity is essentially dedicated to SMI and financing
social economy partners, with no investment banking business. Next is the Credit Mutuel
group, whose results have clearly been affected by the crisis, but mainly not because of direct
risk exposure to the sub-primes and structured securitization products, but rather because of
an insufficiently diversified exposure to other financial institutions that have collapsed. The
choice made by the group of no listed apex and no centralized holding subsidiary has
probably played a stabilizing role in this context. The three other groups (Credit Agricole on
one hand, and Caisse d’Epargne and Banque Populaire on the other) have some key features
in common that speak of a high degree of hybridization. All of them have experienced
significant losses in their investment banking subsidiaries. Nevertheless, the more diversified
model of Credit Agricole has given this group greater resilience to the crisis than the other
two, and helped to compensate for losses at group level.
Above all, the crisis has raised questions about the control of subsidiaries by the
cooperative network. Because of the complexity in how the groups are organized, the
membership exerts less and less effective control on the strategic choices and on the business
model of these subsidiaries, but the crisis has shown that this is likely to play a part in the
risky behaviour of managers.
The second lesson that has to be drawn is that being organized as a cooperative group
is not synonymous with being protected from sanctions by various external discipline
mechanisms, for instance in the case of risk-taking. And these sanctions are likely to have
consequences, not only on the listed vehicles of the group, but also on the cooperative bank
network itself, as described below.

3.2.2 Financial market discipline 21

As soon as there is a listed vehicle in the cooperative group, financial market discipline comes
into play, being manifested in changes in stock price. The effects of market discipline depend
on two variables: the way stockholders estimate the risk level, and the importance and the role
ascribed to the listed company.

21
Most of the information mentioned in this part of the study comes from financial press articles (Les Echos) and from the
bank reference documents and annual reports.
 20

As an example, the decrease in stock price of CIC, the listed vehicle of the Credit
Mutuel group, was in line with the importance of the risk-taking and the losses. Although the
sanction was effective 22, it did not threaten the whole group, because CIC is only a subsidiary
of a holding owned by a regional cooperative, Caisse de Crédit Mutuel, and its position in the
organizational chart is not as important as it is for Credit Agricole and Caisse d’Epargne/
Banque Populaire.
In contrast, in the case of Credit Agricole the apex itself is listed: CASA. The losses
experienced by CALYON, its investment-banking subsidiary, pushed the stock price
downwards strongly23. Capital requirements under Basel 2 regulations increased because the
ratings of counterparties and the structured products and derivatives CALYON was involved
in decreased, and the market value of losses had to be re-evaluated due to new accounting
standard rules. Consequently, the central institution needed capital, and CASA had to issue
new stocks on the financial market. Due to the “solidarity principle” within the group, and
because the regional cooperative banks had to maintain a majority interest in CASA capital,
they had to subscribe to this new share issue using some of their accumulated reserve.
The best illustration of the consequences of hybridization and the existence of a listed
vehicle, however, is to be found in the case of Natixis. As mentioned previously, Natixis is a
holding subsidiary owned equally by the central institutions of the Caisse d’Epargne (CNCE)
and Banque Populaire (BFBP) groups, and is listed on the French financial market. Natixis,
with a large part of its business in investment banking, suffered huge losses, not only because
of excessive exposure to derivatives and securitization structured products, but also because it
held a majority interest in the monocline CIFG (whose exposure to the sub-prime risk was
enormous). In response to the sanctions inflicted on Natixis stocks by the market 24, and in
order to avoid the subsidiary’s bankruptcy and to protect stockholders’ interests, the two
central institutions CNCE and BFPB decided to extract CIFG from Natixis, and to recapitalize
it directly. Despite this decision, they also had to recapitalize Natixis, and, because of the
consolidated data, ultimately absorb the majority of the losses. The consequences for the
regional cooperative banks of these two groups have not been neutral: first, they’ve had to use
part of their own reserves to recapitalize CIFG and their central institution, and to preserve the
activity of the holding subsidiary, even though the investment banking activity is clearly not
of benefit to their own members. Second, they’ve faced a commercial and reputation risk

22
The price decreased from 280 EUR in December 2006 to 101 EUR by the end of December 2008.
23
The price decreased from 35 EUR in December 2006 to 5.9 EUR by the end of December 2008.
24
The price decreased by 80%, between the day it was introduced at 19.5 EUR (beginning of December 2006) and December
2008.
 21

because, although the majority of the losses were associated with Natixis activities, the
commercial brands of Banque Populaire and Caisse d’Epargne have been damaged by their
association with this collapse, especially in the case of customers or members who had been
invited by the commercial teams within the cooperative network to become Natixis
stockholders 25. Third, the cooperative banks have had to accept the consequences of a lower
credit rating, and restructurings in the two groups (see below).

3.2.3 Other external discipline mechanisms

Besides the market itself, the financial crisis has revealed the existence of other external
discipline mechanisms that are likely to penalize cooperative banking groups for excessive
risk-taking.
The first is the actions of the rating agencies. A fall in the rating of a cooperative
banking group is likely to spread its effect over all components of the group, and especially
the cooperative banks themselves, even though they have not been directly involved in any
risk-taking activity. As soon as the central body is rated and issues bonds on the market on
behalf of the cooperative banks, an increase in its estimated credit risk may have
consequences on the rating of the group itself, and therefore, on the cost of financing the
cooperative banks. Of course, one could point out that banks around the world have seen their
ratings downgraded. Nevertheless, despite the high capital ratio (especially tier 1) of the
French cooperative groups, and despite their internal solidarity mechanism, they could not
avoid sanctions from the rating agencies (see Appendix 9). Moreover, whereas some plc
banks have kept a AA or equivalent rating (i.e, BNP-Paribas), and have not been downgraded,
Caisse d’Epargne and Banque Populaire Group have been severely sanctioned (from AA to
A+) by Standard & Poor’s and Fitch, mainly because of their investment banking results and
Natixis losses. However, there is no strong evidence for a link between the degree of
hybridization of the group and the rating, for Credit Agricole and Credit Mutuel have succeed
in keeping a good rating, whereas their structures and organizations are quite different.
The facts also illustrate that, despite the protection conferred by the cooperative legal
form, which prevents the apex or holding subsidiary being threatened by a takeover bid or
being absorbed by another bank 26, another external discipline mechanism has played an
important role, has induced restructurings, and has even driven change in internal governance

25
A complaint was lodged against the two groups by minority shareholders.
26
Cf. for example FORTIS, acquired by BNP-PARIBAS.
 22

mechanisms. According to financial press analyses, this has been the case for two of the
French cooperative banking groups: Caisse d’Epargne and Banque Populaire. These two
groups were weakened by the collapse of Natixis, other internal conflicts and losses on the
market 27, and had to ask for public financial support. The French government would almost
certainly have placed conditions on any help offered, in order to accelerate significant
restructuring in the groups 28, and including a change in direction. The consequences for the
organization and for the independence of the cooperative banks within the “new group” are
not yet known explicitly, but there is enough evidence to suggest that the importance given to
French State interest has increased 29, and that, contrary to standard cooperative rules, the
president of the group will not come from a previous circle of cooperative administrators or
from the Banque Populaire or Caisse d’Epargne membership.

CONCLUSION

To conclude, let us come back to Agency Theory, the theoretical background used as a
reference tool by the mainstream of Anglo-American authors to justify the relatively poor
efficiency of cooperative banks. We have shown that, until the crisis, these arguments did not
hold in the French context and that the cooperative banks have been able to adapt, to resist a
threatening regulatory and accounting environment, and to withstand comparison with plc
banks in terms of efficiency. Nevertheless, the recent financial crisis has highlighted the fact
that the hybridization of the original cooperative model was not neutral, and the impact of the
crisis has spread from investment banking activities and subsidiaries held by the cooperative
networks to impact on all the stakeholders within the groups (shareholders, managers,
customers and members). Therefore, as mentioned by the Anglo-American authors, the
question of control within the group itself, and especially that of managers and leading
administrators, has become an important issue. In light of the crisis, it seems one of the most
important issues for the future is the control on the subsidiaries of the group, and the possible
role of the members in the strategic choices of the group, and not only of the cooperative

27
The French financial press articles (www.lesechos.fr) state internal debates among the regional banks concerning the
burden of Natixis losses and the responsibilities that have to be assumed; they also mention trading loss of 700 millions of
EUR, on the equity capital of the CNCE.
28
The ongoing restructuring process is characterized by a merger of the two central bodies (CNCE and BFBP), has lead to a
redefinition and a restriction of the scope of activity of Natixis, and a cut off in employment; the two cooperative networks
and commercial brands (Banque Populaire and Caisse d’Epargne) still coexist, but some worries have come out in the
regional banks, after a planned redundancy scheme was announced by the CNCE.
29
At the moment, the French State has taken a 20% interest in the capital of the new body of the group (5 billions of EUR,
either as preferred shares, either as super-subordinated debt).
 23

network. Despite the fact that Agency Theory is not relevant in itself to explain the
hybridization of the cooperative banks, and that some of the explanations have to be searched
in neo-instutitionalism references and analyses (DE SERRES, JAEGER & ORY, 2010).
Nevertheless, it allows to point out some of the potential interest conflicts that have to be
solved in the future: to what extent is the original cooperative model able to survive in such
hybrid groups? Is the stability and resilience to financial shocks that could be expected thanks
to the natural activity of the cooperative banks, and by reference to the values affirmed by the
ICA 30, threatened by an overexposure to risk and the introduction of the shareholder value
logic?

30
ICA: International Cooperative Alliance; cf. for instance to DRAPERI (2005) for the terms of these principles.
 24

References

AKELLA S.R., GREENBAUM S.I. (1988), « Savings and Loans Ownership Structure and Expense-
Preference », Journal of Banking and Finance, n°12, p. 419-437.
ALBERT M. (1991), Capitalisme contre capitalisme, Le Seuil, Paris
BERGER, A.N., MESTER L.J. (1997), « Inside the Black Box: What Explains Differences in the
Efficiency of Financial Institutions », Journal of Banking and Finance, n°21, p. 895-947.
CHARREAUX, G. (1997a), Vers une théorie du gouvernement des entreprises, in CHARREAUX ed.,
« Le gouvernement des entreprises », Paris, Economica.
CHARREAUX G. (1997b), Gouvernement de l’entreprise, Encyclopédie de Gestion, P. JOFFRE et Y.
SIMON ed., p 1650-1662.
CHATTERJEE S., (2003), Enron : Incremental Descent into Bankruptcy : a Strategic and
Organisational Analysis, Long Range Planning, vol. 36, 2, April, p 133-149.
DETILLEUX J.C., NAETT C. (2005), "Les coopératives face aux normes comptables internationales,
le cas de l'IAS 32", Revue Internationale de l'Economie Sociale (RECMA), n°295, février
DRAPERI J.F (2005), L’économie sociale : utopies, pratiques, principes, Presses de l’Economie
Sociale, 125 p.
EACB (2006), « EACB Resolution on IAS 32 », Groupement Européen des Banques Coopératives,
Brussels, december.
EACB (2007), « The Discussion on Loan-Loss Absorption Approach, memo, GEBC, Brussels, 23
avril
FAMA E.F., JENSEN M.C. (1983), « Agency Problem and Residuals Claims », Journal of Law and
Economics, n° 26, p. 327-349.
FONTEYNE, W. (2007), “Cooperative Banks in Europe: policy issues”, IMF Working Paper, July.
GURTNER E., JAEGER M., ORY J.N. (2004), « Cooperative Status in the Banking Sector,
Efficiency and Growth Strategy: the French Situation », Newsletter, n°187, EFMA, p. 44-47.
HANSMANN H. (1985), « The Organization of Insurance Companies: Mutual versus Stocks »,
Journal of Law, economics and Organization, vol. 1, n° 1, p. 125-153.
HANSMANN H. (1996), "The ownership of Enterprise", Harvard University Press
HANSMANN H., KRAAKMAN, R. (2001) "The End of History for Corporate Law" Harvard
Discussion Paper n°280
HART O, MOORE J. (1990), “Property rights and the nature of the firm”, Journal of Political
Economy, n° 98, December, p 1119-1158
HIEZ D. (2005), "Les instruments de fonds propres des coopératives: vingt ans d'évolution
législative", RECMA, Revue internationale de l'Economie sociale, n°295, février 2005
JENSEN M.C (2001), Value Maximization, Stakeholder Theory, and the Corporate Objective
Function, Working Paper, Harvard Business School, and Amos Tuck School at Dartmouth College.
JENSEN M.C, MECKLING W.H. (1976), Theory of the firm: Managerial Behaviour, Agency Costs
and Ownership Structure, Journal of Financial Economics, October, p 305-60
LAMM-TENNANT J., STARKS L.C. (1993), « Stock versus Mutual Ownership Structures: the Risk
Implications », The Journal of Business, vol. 66, n°1, p. 29-46.
"Les banques coopératives en France : le défi de la performance et de la solidarité", RICHEZ-
BATTESTI et GIANFALDONI dir., L’Harmattan, Paris, 2006.
 25

MAYERS D., SMITH C. (1994), « Managerial Discretion, Regulation and Stock Insurance Ownership
Structure », the Journal of Risk and Insurance, vol. 61, n° 4, p. 638-655.
MESTER L.J (1992), « Traditional and non traditional banking », Journal of Banking and Finance,
vol. 16, n°3, p. 545-566
MESTER L.J (1993), « Efficiency in the Savings and Loans Industry », Journal of Banking and
Finance, April, vol. 17, n° 3, p. 267-286.
O'SULLIVAN M. (2002), « Le rôle du marché boursier dans les systèmes nationaux de
gouvernance », Revue Française de Gestion, novembre-décembre, n° 141.
ORY J.N, JAEGER M., GURTNER (2006 a) « La banque à forme coopérative peut-elle durablement
soutenir la compétition avec la banque S.A ? », Finance Contrôle Stratégie, juin, p121-58
ORY J.N, GURTNER E., JAEGER M. (2006 b) «The Challenges of the recent Changes in the French
cooperative banking Groups», Revue Internationale de l’Economie Sociale (RECMA), October, p 43-
59
PERRET, V. JOSSERAND E. (2003), “Pratiques organisationnelles du Paradoxe”, in PERRET V.,
JOSSERAND E (dir.), Le paradoxe : penser et gérer autrement les organisations, Ellipes, p 165-187.
RASMUSEN E. (1988), « Mutual banks and stock banks », Journal of Law and Economics, n°31, p.
395-421
ROE M. (1994), Strong Managers, Weak Owners; the Political Roots of American corporate Finance,
Princeton University Press
SMITH B.D. (1984), « A Theoretic Framework for the Analysis of Credit Union Decision Making»,
Journal of Finance, September, p. 1155-1168
SOULAGE F. (2000), “Les fonds propres des coopératives : analyse et propositions”, ESFIN-IDES,
mai
TURNBULL, MACQUARIE (2000) "The competitive advantages of stakeholder mutuals", Working
Paper, 12th Annual Meeting on Socio-Economics, London School of Economics, Network H, Markets
and Institutions
WEISS M. (2004), Organizational Design in the Banking Industry: a comparative institutional
Analysis of the German Cooperative Banking Groups, WP, Goethe Institut, Frankfurt.
WILLIAMSON O.E. (1983), « Organization Form, Residual Claimants, and Corporate Control »,
Journal of Law and Economics, n° 36, p. 351-366
 26

Appendix 1

List of the French cooperative banking networks or groups:

Banque Populaire
Caisse d’Epargne (since 1999: cooperative status)
Credit Agricole
Credit Cooperatif
Credit Mutuel
Credit Maritime Mutuel

List of the 30 European plc banking groups (total of assets criterion):

Intesa Sanpaolo (Italy)
Bayerische Hypo-und Vereinsbank AG (Germany)
SANPAOLO IMI (Italy)
UniCredit Italiano SpA (Italy)
Commerzbank AG (Germany)
Société Générale (France)
BNP Paribas (France)
Banco Bilbao Vizcaya Argentaria SA (Spain)
Dresdner Bank AG (Germany)
Santander Central Hispano Group-Banco Santander Central Hispano (Spain)
KBC Bank NV (Belgium)
Deutsche Bank AG (Germany)
HSBC Bank plc (United Kingdom)
Deutsche Postbank AG (Germany)
ING Bank NV (Netherlands)
Royal Bank of Scotland Plc (The) (United Kingdom)
Barclays Bank Plc (United Kingdom)
Lloyds TSB Bank Plc (United Kingdom)
Dexia Bank-Dexia Bank Belgium (Belgium)
Danske Bank A/S (Denmark)
ABN Amro Holding NV (Netherlands)
Bank of Scotland (United Kingdom)
Skandinaviska Enskilda Banken AB (Slovenia)
Svenska Handelsbanken (Slovenia)
Dexia Crédit Local SA (France/Belgium)
Goldman Sachs International (United Kingdom)
UBS Limited (United Kingdom)
Fortis (Belgium)
HBOS Treasury Services Plc (United Kingdom)
Depfa Bank Plc (Ireland)
 27

Appendix 2:
List of the efficiency ratios used in the Wilcoxon tests

↓denotes the fact that a good performance is a low ratio

↑denotes the fact that a good performance is a high ratio

Credit risk ratios:

- Loan loss provisions / net interest revenues: ↓
- Loan loss reserves / impaired loans: ↑

Commercial ratios:
- Net interest margin ↑
- Net interest revenues / average assets: ↑

Operational efficiency ratios :

- Personal expenses / average assets 31: ↓
- Cost to income ratio: ↓
- Pre-tax operating income/average assets: ↑

Equity capital ratios:

- Equity / total assets : ↑

Financial efficiency ratios:

- Return on average equity (ROE) :↑
- Return on average assets (ROA) ↑

31
Average between the assets at the end and at the beginning of the year.
 28

Appendix 3: Methodology of the Wilcoxon test

Example of a Wilcoxon test performed on the ratio : Net interest margin/ average assets (NIM)
% %
année NIM NIM ∆ = NIM FcopB – ∆ sign Rank of ∆
FcopB FplcBG NIM FplcBG
1995 2,08 1,61 0,47 + 4
1996 1,54 1,4 0,14 + 1
1997 1,56 1,34 0,22 + 2
1998 0,98 1,24 -0,26 - 3
1999 1,9 0,79 1,11 + 10
2000 1,75 0,82 0,93 + 7
2001 1,78 0,78 1 + 8
2002 1,75 0,96 0,79 + 5
2003 1,83 0,97 0,86 + 6
2004 1,87 0,85 1,02 + 9
W+ =52

The method is the following: Every year the difference (∆) of the observed ratio for French
Coop. Banks (FcopB) and French plc banking groups (FplcBG) is calculated. Thus, a sample
of differences, positive or negative, is available.
A rank is associated too each difference (from 1 to 10, considering the period), and is ordered
according to an increasing absolute value criterion. Under the null hypothesis H(0) of equal
efficiency between FcopB and FplcBG, the sum of the positive ranks W+ should
approximately be the same as the sum of the negative ranks. By contrast, under the H(1)
hypothesis of higher efficiency for the FcopB, W+ is expected to be larger than W- .
The test criterion is the sum of the positive ranked differences, W+ , whose distribution is
written in the Wilcoxon statistical table, under H(0). Thus, for ten elements (years), the H(0)
hypothesis of equal efficiency can be rejected as soon as W+ exceeds 45 (5% risk level). The
critical p-value attached to this number (45) can be read in the Wilcoxon table: 4,2%.
The result of the test, in this example is : Σ W+ (Bco) = 54 ; 52 > 45 (critical value)
→ H(0) is rejected
conclusion : higher efficiency for the FcopB than the (FcopB > FplcBG)
 29

Appendix 3

Score methodology applied to the cooperative networks (and groups):

Once the Wilcoxon test enables to know which cooperative network/group exhibits a better
result than another one, a global classifying by indicator is operated, and a score is given to
each of the networks/groups, as explained below:
if a network/group is performing better than all the three others on an indicator he gets a score
equal to 4 (as they are four considered networks); the network that is performing better than
two others gets the second place in the classifying with a score equal to 3…etc…

Example of Wilcoxon test applied to the four cooperative networks, according to the Net
Interest Margin criterion:

Available period Wilcoxon Critical value Rejection of the Résults

Test (5%risk level) H(0) equality
Criterion : efficiency
W+

Net Interest 1995-2004 55 45 Yes CE<CA

Margin
1995-2004 55 45 Yes CE>CM
1995-2004 55 45 Yes CE<BP
1995-2004 55 45 Yes CM<CA
1995-2004 55 45 Yes CA<BP
1995-2004 55 45 Yes CM<BP
The sign < denotes the fact that for the considered indicator, the first network exhibits a statistically lower
performance than the second one (and conversely for the > sign)

We can notice that the network that performs better than all the others is Banque Populaire
(BP); it is followed by Credit Agricole (CA), then Caisse d’Epargne (CE) and at last Credit
Mutuel (CM). According to this score methodology, Banque Populaire gets 4, Credit Agricole
gets 3, Caisse d’Epargne gets 2 and Credit Mutuel gets 1.

Then, the same methodology is replicated for each of the indicators. As we have classified the
indicators into five different categories (credit risk, commercial efficiency, financial
efficiency, equity capital, operational efficiency), a simple score-mean is calculated for eachof
these categories.
Finally, a global simple mean based upon all the different categories is calculated. This last
one can be seen as a global efficiency score of a cooperative network/group, all of the
efficiency categories taken into account with the same weight coefficient.
 30

Appendix 4: Wilcoxon tests results and efficiency scores of the French cooperative
networks and groups versus French and European plc groups (1995-2004)

FRENCH COOP. GROUPS AND NETWORKS

versus PLC BANKS

FCOPB EPLC FPLC FCOPBG

Credit risk ratio
LLR/IL 3 3 3 1
Mean 3 3 3 1
Capital equity
ratio
EQ/TA 4 2,5 1 2,5
Mean 4 2,5 1 2,5
Commercial
performance
NIR/AA 3,5 3,5 1,5 1,5
NIM
Mean 3,5 3,5 1,5 1,5
Operational
performance
COST/INC 2 4 2 2
PTOI/AA 3,5 3,5 1,5 1,5
PERS EXP/TA 1 2 4 3
Mean 2,17 3,17 2,50 2,17
Financial
performance
ROA 3,5 3,5 2 1
ROE 1,5 3,5 3,5 1,5
Mean 2,5 3,5 2,75 1,25

Total Mean 3,03 3,13 2,15 1,68

LEGEND :
FCOPB: French Cooperative Banks (network only, excluding subsidiaries)
FCOPBG: French Cooperative Banking Groups (including subsidiaries’ activity)
FPLC: French plc banks
EPLC: European plc banks

Efficiency indicators :
LLR/IL: Loan Loss Reserve/ Impaired Loans
EQ/TA: Equity/Total Assets
NIR/ AA: Net Interest Revenues/ Average Assets
NIM: Net interest Margin
COST/INC: Cost to Income
PTOI/ Average Assets: Pre Tax Operating Income/ Average Assets
PERS EXP/TA: Personnel Expenses/ Total Assets
 31

Appendix 5: Wilcoxon test results and efficiency scores

of the French cooperative networks (1995-2004)

FRENCH COOPERATIVE BANKING NETWORKS

CE BP CA CM
Credit risk ratio
LLR/GL 4 1.5 1.5 3
LLP/NIR 4 1.5 1.5 3
Mean 4 1.5 1.5 3
Capital equity
ratio
EQ/TA 1.5 3 4 1.5
Mean 1.5 3 4 1.5
Commercial
performance
NIR/AA 2 4 3 1
NIM 2 4 3 1
OOI/AA 2 3.5 3.5 1
Mean 2 3.8 3.2 1
Operational
efficiency
COST/INC 2 3 4 1
PTOI/AA 2 3,5 3,5 1
PERS EXP/TA 3 1 2 4
Mean 2,3 2,5 3,2 2,0
Financial
performance
ROAA 2 3,5 3,5 1
ROE 2 4 3 1
Mean 2 3,75 3,25 1

Total mean 2,4 2.9 2,8 1,7

LEGEND:
CE: Caisse d’Epargne
BP: Banque Populaire
CA: Credit Agricole
CM: Credit Mutuel

Efficiency indicators:
LLR/GL: Loan Loss Reserve/ Gross Loans
EQ/TA: Equity/Total Assets
NIR/ AA: Net Interest Revenues/ Average Assets
NIM: Net interest Margin
OOI/AA : Other Operating Income/ Average Assets
COST/INC: Cost to Income
PTOI/AA: Pre Tax Operating Income/ Average Assets
PERS EXP/TA: Personnel Expenses/ Total Assets
 32

Appendix 6: Wilcoxon test results and efficiency scores

of the French cooperative groups (1995-2004)

FRENCH COOPERATIVE BANKING GROUPS

CE BP CA CM
Credit risk ratio
LLR/GL 4 2 2 2
Mean 4 2 2 2
Capital equity ratio
EQ/TA 1,5 1,5 3 4
Mean 1,5 1,5 3 4
commercial performance
NIR/AA 1,5 3,5 1,5 3,5
NIM 1,5 3,5 1,5 3,5
Mean 1,5 3,5 1,5 3,5
Operational performance
COST/INC 1 3 3 3
PTOI/AA 2 4 2 1
PERS EXP/TA 3,5 1,5 1,5 3,5
Mean 2,17 2,83 2,17 2,50
Financial performance
ROAA 1,5 3,5 1,5 3,5
ROE 3 4 2 1
Mean 2,25 3,75 1,75 2,25

Total mean 2,28 2,72 2,08 2,85

LEGEND :
CE: Caisse d’Epargne
BP: Banque Populaire
CA: Credit Agricole
CM: Credit Mutuel

Efficiency indicators:
LLR/GL: Loan Loss Reserve/ Gross Loans
EQ/TA: Equity/Total Assets
NIR/ AA: Net Interest Revenues/ Average Assets
NIM: Net interest Margin
OOI/AA: Other Operating Income/Average Assets
COST/INC: Cost to Income
PTOI/ Average Assets: Pre Tax Operating Income/ Average Assets
PERS EXP/TA: Personnel Expenses/ Total Assets
 33

Appendix 7: efficiency indicators, and growth rates from 2006 to 2008

(source: annual reports – Millions EUR when not precised)

BNP-Paribas Societe Generale Caisse Epargne Banque Populaire

2008 2007 2006 2008 2007 2006 2008 2007 2006 2008 2007 2006
BNP 32 group
(M€) 27376 31037 27943 BNP group (M€) 21866 21923 22417 BNP group (M€) 8400 9800 11000 BNP group (M€) 7253 7445 8083

BNP Retail bank 5943 6000 5850 BNP Retail bank 7191 7058 6883 BNP Retail bank 5234 5563 5815 BNP Retail bank 5698 5839 5578
33
BNP CIB 4973 8293 7859 BNP CIB 4017 4522 6998 BNP CIB -213 635 1211 BNP CIB 1027 1923 2516

OGR 34 group 8976 12273 10878 OGR GROUP 6338 7618 8714 OGR GROUP -200 1500 2400 OGR GROUP 968 1537 2750
OGR Retail
OGR Retail bank 1960 2050 2039 bank 2513 2492 2383 OGR Retail bank 1137 1550 1776 OGR Retail bank 1895 2181 2047

OGR CIB 1262 3508 3462 OGR CIB 539 1097 3108 OGR CIB -868 -56 481 OGR CIB -764 92 799

COI 35 GROUP 67,20 60,50 61,10 COI GROUP 71 65 61,10 COI GROUP 102,38 84,00 74,9 COI GROUP 86,70 79,36 65,98
COI Retail bank 67,02 65,83 65,15 COI Retail bank 65 65 65 COI Retail bank 78,00 72,00 69,46 COI Retail bank 66,3 62,7 63,3
COI CIB 74,62 57,70 55,95 COI CIB 87 76 55,60 COI CIB 307,51 108,82 66,9 COI CIB 174,39 95,22 68,24

32
BNP: Banking Net Product
33
CIB: Corporate and Investment Bank
34
OGR: Operational gross Result
35
COI: Cost to Income ratio (%)
 34

BNP-Paribas Societe Generale Caisse Epargne Banque Populaire

36 37
OR GROUP 3224 10548 10095 OR 3683 6713 8035 OR GROUP -1641 1262 2819 OR GROUP -236 1052 2442

OR Retail bank 1757 1892 1886 OR Retail bank 2033 2163 2108 OR Retail bank 964 1432 1682 OR Retail bank 1366 1793 1767

OR CIB -1215 3480 3715 OR CIB 485 1153 3201 OR CIB -1405 -133 322 OR CIB -1414 8 782

RC 38 GROUP -5752 -1725 -783 RC GROUP -2655 -905 -679 RC GROUP -1441 -259 -23 RC GROUP -1204 -485 -308
RC Retail bank -203 -158 -153 RC Retail bank -480 -329 -274 RC Retail bank -173 -118 -94 RC Retail bank -529 -388 -335

RC CIB -2477 -28 -253 RC CIB -1024 -56 -93 RC CIB -537 -77 -4 RC CIB -650 -85 -17

Net Result
Net Result Group 3021 7822 7308 Group 2010 947 5221 Net Result Group -2000 1400 3832 Net Result Group -468 1055 1700
Net Result Net Result Net Result
Retail Bank 1296 1375 1344 Retail Bank 632 923 1136 Retail Bank 1026 1340 1199
Net Result CIB -235 -2221 2340 Net Result CIB -1159 -61 326 Net Result CIB -1375 316 635

GROUP Equity 47511 41686 38555 GROUP Equity 36100 27200 29100 GROUP Equity 16600 20600 20000 GROUP Equity 17700 20200 20400

ROE 39 GROUP
(%) 6,36 18,76 18,95 ROE GROUP 7,39 3,25 ROE GROUP -12,05 6,80 19,16 ROE GROUP -2,64 5,22 8,33

36
OR: Operational result
37
Excluded the consequences of Jerome Kerviel illegal activity ; idem fot Net result CIB.
38
RC:Cost of Risk
39
ROE: Return On Equity (%). So as to give the same meaning to this ratio for each individual, it has been calculated this way: (group share net result at the end of the N-year/group share equity
at the beginning of the N-year).
 35

Appendix 7 bis

Credit Mutuel Credit Agricole

2008 2007 2006 2008 2007 2006
BNP GROUP
BNP GROUP (M€) 8424 10568 10838 (M€) 15956 16768 16187
BNP Global Retail Bank 40 7485 7561 7504 BNP Retail bank 11262 11960 12093
BNPcoop. Network
Retail bank 4619 4664 4695 BNP CIB 1893 2781 5456
41
BNP of CIC 2866 2897 2809 BNP of CALYON 2153 3359 5862
42
BNP CIB -64 1043 1252

OGR 43 GROUP 1747 4057 4495 OGR GROUP 3321 4050 5832
OGR Global Retail Bank 2081 2293 2418 OGR Retail bank 4347 4955 5171
OGR Network Retail bank 1310 1475 1613 OGR CIB -1687 -756 2135
OGR of CIC 771 818 805 CALYON -1738 -470 2288
OGR CIB -440 622 885,1

COI 44 GROUP (%) 79,26 61,61 58,53 COI GROUP 79,186513 75,846851 63,9711
COI of Global Retail
Bank 72,20 69,67 67,78 COI Retail bank 61,40 58,57 57,24
COI of Coop. Network
Retail bank 71,64 68,37 65,64 COI of CIB 189,1 127,2 60,9
COI of CIC 73,10 71,76 71,34 COI of CALYON 79,2 39,4 40,3
COI of CIB 587,50 40,36 29,31

OR 45 GROUP 342 3871 4256 OR GROUP 156 2153 5220

OR Global Retail Bank 1523 2118 2154,8 OR Retail bank 3010 3971 4330
OR of Coop. Network
Retail bank 1050 1405 1455 OR of CIB -2997 -1713 2145
0R of CIC 473 713 700 OR of CALYON -3066 -1430 2297
OR of CIB -1159 615 915,4

RC 46 GROUP -1405 -186 -239 RC GROUP -3165 -1897 -612

RC Global Retail Bank -558 -175 -263,2 RC Retail bank -1337 -984 -841
RC of Coop. Network
Retail bank -260,00 -70,00 -158,20 RC of CIB -1310 -957 -10
RC of CIC -298,00 -105,00 -105 RC of CALYON -1328 -960 9
RC CIB -719 -7 30,3

Net Result Group 440 2730 2946 Net Result Group 1024 4044 4860
Net result of Net Result
Global Retail Bank 1017 1431 1431,3 RetailBanking 581 778 750
Net result of Coop
Network Retail bank 619 857 894,3 Net Result of CIB -1924 -904 1645
Net Result of
Net result of CIC 398 574 537 CALYON -1540 -714 1738
Net Result of CIB -736 496 726,3
GROUP Equity 26442 23983 20530 GROUP Equity 40691 34319 30047
ROE GROUP (%) 1,66 11,38 14,35 ROE GROUP 2,52 11,78 16,17

40
“Global retail Bank” includes not only the results of the cooperative Network Retail Bank but also those of the plc banks
which belong to the CIC subsidiary
41
CIC is the main plc holding subsidiary of Credit Mutuel
42
CIB : corporate and investment banking
43
OGR : Operational Gross Result
44 COI : Cost to Income ratio (%)
45
OR : Operational Result
46
RC : Cost of Risk
36
 37

Appendix 8: Efficiency ratios

(Bankscope data base, French Central Bank, Annual Reports)

2008 1992-1996
(Group perimeter of activity) (Aggregated Data by legal status)

SG BNP BP CA CE CM plc cooperative banks

banks (networks)
Efficiency Ratios
Return on Equity (%) 7,69 6,83 -2,09 4,19 -9,81 1,7 -2,18 6,24
Return on Average Assets (%) 0,25 0,18 -0,11 0,18 -0,32 0,08 -0,07 0,32
Net Interest Margin (%) 0,83 0,81 1,16 1,28 0,83 0,75 1,61 3,29
Operational Gross Result/Banking Net
Product (%) 28,98 32,78 13,34 20,81 -2,45 20.7 22 32
Operational Result/ Banking Net Product -
(%) 16,84 11,77 3,25 0,97 -19,56 4,06 -0.2 22.6
Net result/ Banking Net Product (%) -
9,19 11,03 6,45 6,41 -23,96 5.22 -4 9,9
Cost to Income (%) 71,01 67,2 86,7 75,16 102,43 80,52 78,09 69,5
Equity Ratios
Equity/Total Assets (%) 4,85 3,72 6,78 5,64 4,94 5,77 3,28 5,11
Equity/Debt 3,8 2,42 5,36 4,1 2,98 4,57 ND ND
Solvency Ratio (Basle 2) (%) 11,2 11,1 9,4 9,9 9,6 9,5 ND ND
Risk
Cost of Risk /Banking Net Product (%) 12,14 21,01 20.46 19,84 17,14 16,68 22.2 9.4
Loan Loss Provision / Net Interest
Revenues (%) 31,56 26,76 29,89 23,98 30,14 35,39 ND ND
Impaired Loans/ Gross Loans (%) 3,74 4,49 3,7 2,87 1,65 2,97 ND ND
General Loan Loss Reserve /
Impaired Loans (%) 62,77 62,57 70,04 89,55 72,37 69,14 ND ND
 38

Appendix 9:
Evolution of the ratings of the French cooperative and plc banking groups
(2006-2008)

2006 2007 2008

Caisse S&P: AA S&P : AA S&P: A+
Epargne Fitch: AA Fitch: AA- Fitch: A+
Moodys’: Aa2 Moodys’: Aa2 Moodys’: Aa3
Banque S&P: AA S&P : AA- S&P: A+
Populaire Fitch: AA Fitch: n.a Fitch: A+
Moodys’: Aa2 Moodys’: Aa2 Moodys’: Aa3
Credit S&P: AA S&P : AA- S&P: AA-
Agricole Fitch: AA Fitch: AA Fitch: AA-
(CASA) Moodys’: Aa2 Moodys’: Aa1 Moodys’: Aa1,
negative watch
Crédit S&P: AA S&P: AA- S&P: n.a
Mutuel Fitch: AA Fitch: AA- Fitch: AA-
(BFCM) Moodys’: Aa3, Moodys’: Aa3, Moodys’: Aa3,
positive watch positive watch positive watch
BNP S&P: AA, S&P: AA+ S&P: AA,
Positive watch Negative watch
Fitch: AA Fitch: AA Fitch: AA,
Moodys’: Aa2 Moodys’: Aa2 Negative watch
Moodys’: Aa2,
negative watch
Societe S&P: n.a S&P: n.a S&P: AA,
Generale negative watch
Fitch: n.a Fitch: AA Fitch: AA-
Moodys’: n.a Moodys’: n.a Moody’s: n.a