Palen by shuifanglj

VIEWS: 1 PAGES: 35

									CORPORATE GOVERNANCE AND EFFICIENCY:
 EVIDENCE FROM SPANISH SAVINGS BANKS

               (Preliminary version)


             Valentín Azofra Palenzuela
          Dpto. Economía y Admón. Empresas
        Avda. Valle de Esgueva, 6 –47011 Valladolid
                  Tlf: 34 – 983 – 423333
                  Fax: 34 – 983 – 423899
                Email: vazofra@eco.uva.es

            Marcos Santamaría Mariscal
          Dpto. Economía y Admón. Empresas
             C/ Parralillos, s/n – 09001 Burgos
             Tlf: 34 – 947 – 259040 / 258968
                  Fax: 34 – 947 – 258960
                  Email: msanta@ubu.es
                                         -2-


    CORPORATE GOVERNANCE AND EFFICIENCY:
     EVIDENCE FROM SPANISH SAVINGS BANKS

ABSTRACT: This paper analyses the relation between the composition of the
governing bodies and the efficiency of Spanish savings banks in the year 1999.
Using Tobit models, the results of the study demonstrate the existence of a negative
relation between the participation of local and regional governments on the board of
directors and cost efficiency of savings banks. This result is maintained in both cases,
when we use the operating expenses to ordinary income ratio as a proxy for the cost
efficiency and we use an efficient frontier non parametric model called DEA (Data
Envelopment Analysis). Likewise, a positive relation is found between the
participation of politicians on the board and the channelling of more financial
resources towards social and cultural projects.


KEY WORDS: Agency theory; corporate governance; board of directors; savings
banks; efficiency.
                                         -3-




         CORPORATE GOVERNANCE AND EFFICIENCY:
          EVIDENCE FROM SPANISH SAVINGS BANKS


0- INTRODUCTION

      In the last few years the debate on the effectiveness of mechanisms of
corporate governance have generated an important tide of research aimed at
clarifying the relation existing between corporate governance and the value of the
firm. From a limited agency approach, that is, considering only the agency relation
between managers and providers of financial resources, a greater effectiveness in
monitoring management through different control mechanisms will lead to a
reduction in agency costs which will benefit the value of the organisation. The scarce
literature in financial firms has revealed that ownership structure, market for
corporate control, banking regulation and the board of directors are the most relevant
control mechanisms.


      Spanish savings banks are financial intermediaries characterised by the non
existence of alienable property rights and by an assignation of control rights through
laws. This particular organisational form determines the importance of internal
control mechanisms to solve the agency problem. In savings banks this problem
refers to how the different stakeholders, especially depositors as the principal
contributors of funds, can ensure that the resources they provide are used effectively.
The double role played by regional governments, by acting simultaneously as
regulators and participants in the governance of savings banks, determines that
control rights are assigned mainly to the political group to the detriment of the rights
of other groups. The consequence is that the public authorities have a direct and
significant influence on investment and financing decisions of savings banks, and on
                                         -4-


the funds which are earmarked for social and cultural projects. The absence of
effective governance mechanisms, especially through the market for corporate
control and/or banking regulation, means that savings banks can adopt some
operative and strategic decisions guided more by political interests than by economic
ones. In the end, these decisions will negatively affects the efficiency of the firm. To
sum up, this paper principally analyses the influence of local and regional
governments in the board of directors of Spanish savings banks on their cost
efficiency calculated, alternatively, through an efficient frontier model and the ratio
operating expenses divided by ordinary income.
      The paper is divided into four sections. The first sets out the basis of the
problem of corporate governance of financial firms, highlighting the concentration of
ownership, regulation and the board of directors as the most relevant mechanisms.
The second section applies the prior theoretical framework to the study of the
specific agency problem in Spanish savings banks and a set of hypotheses are posed
which are tested in the third section. Finally in section four we present the main
conclusions.



1- CORPORATE GOVERNANCE

      From the theoretical and conceptual framework provided by agency theory
(Jensen and Meckling, 1976) it is recognised that the basic problem of the
organisations where there is a separation between the ownership and the control of
resources centres on the difficulties faced by the owners to ensure that the funds they
contribute are not wasted by management on unprofitable projects (Shleifer and
Vishny, 1997). However, this perspective, focused on a single agency relation
between shareholders and management, is recognised as being too restrictive when
analysing the participation in the firm of other agents such as employees, customers,
creditors or government. In this context, the agency problem takes on a more global
character as a result of coming together in the framework of the firm of multiple
contractual relations amongst different stakeholders. These      stakeholders desire to
maximise their utility even to the detriment of the firm‟s global value. The solution
to this problem requires the development of an optimal system of institutional
                                          -5-


agreements and relations which allow for an overall reduction in the agency costs
derived from opportunist behaviour (Fernández, 2001). Thus explained, the
governance of the firm, or corporate governance, would serve as model to solve
agency conflicts amongst stakeholders. The firm‟s results will depend on its
annulment, or perhaps minimisation. From this perspective, essential aspects are
included such as what role the managers played as a central piece in contractual
relations, how the ownership structure influences the control of managerial
discretion, how the design of governing bodies affects the supervision and resolution
of these problems, and more recently, whether it is feasible to share out decision
rights and control amongst the different stakeholders within the board of directors
(Rajan and Zingales, 1998).

      Nevertheless, the financial literature on the governance of the firm has
traditionally focused its attention on the contractual relations between providers of
funds (owners and creditors) and management. The supposition, which underlies this
relation, is that greater effectiveness in the monitoring and control of management
through different mechanisms of governance would imply a reduction in agency
costs which finally leading to the greater value of the firm. Very little attention has
been paid to the governance of financial firms. However, the especially distinctive
characteristics of financial intermediaries (subject to intense regulation and the key
importance of their principal creditors-depositors) together with the essential role
they play in the economy require the special analysis of their corporate governance.
Despite this, the study of financial firms has paid specific attention to aspects related
with the effectiveness of the market for corporate control (Brickley and James, 1987;
Schranz, 1993; Hubbard and Palia, 1995), the banking regulation (Freixas, 1998;
Prowse, 1997 a and b; Ciancanelly and Reyes, 2000), the ownership structure (Allen
and Cebenoyan, 1991; Gorton and Rosen, 1995; Hirschey, 1999; Anderson and
Fraser, 2000;) or the composition of the board of directors (Pi and Timme, 1993;
Simpson and Gleason, 1999).
                                           -6-


      Our paper focuses its attention on banking regulation, concentration of
ownership and board of directors as relevant mechanisms in the supervision of
managerial behaviour.
      It is a fact that financial firms operate under completely different legal and
regulatory settings than non-financial companies. The existence of assets with risk
which can be acquired by small uninformed investors (depositors) together with the
important negative externalities implied in the bankruptcy of these firms for the
financial system of a country require a special regulatory framework through ex–ante
and ex-post supervision (Freixas and Rochet, 1999). Restriction in merger processes
and acquisitions, limitations to entry of new firms or to creation of new branches,
capital requirements or the establishment of a deposit insurance are some of the
measures adopted by public authorities to try to guarantee the correct behaviour of
these firms. The importance of banking regulation as a key instrument in the control
of managerial opportunism has already been underlined by Prowse (1997 a and b)
and Ciancanelli and Reyes (2000), who recognise it as the principal mechanism of
corporate governance in financial firms.


      The literature on corporate governance shows that the concentration of
ownership through the presence of a large shareholder may help to mitigate the
agency problem between shareholders and management. As Azofra (2000) points
out, an organisation which it is possible to transfer title deeds without loss of value,
and where there is an important dispersion of capital amongst multiple shareholders,
these do not have sufficient incentives to control the behaviour of managers -the
traditional free rider problem- (Grossman and Hart, 1980). On the other hand the
existence of a majority shareholder with the capacity to control and fire managers
may reduce possible opportunist behaviour. Thus, Hannan and Mavinga (1980) find
evidence of a positive relation between banks with dispersed ownership and the
expenditure on salary of their managers. Akella and Greenbaum (1988) confirm that
North American mutual savings banks, as the maximum example of the extreme
dispersion of ownership, tend to expand their deposits and loans beyond the optimum
level of profit maximisation, which generates high agency costs. Glassman and
Rhoades (1980) verify, by a sample of North American banks, the existence of a
                                          -7-


positive relation between concentration of ownership and better profit rates.
However, concentration of ownership is not without its problems, especially when
the majority shareholder uses his influence to pursue his own goals at the expense of
the interests of the rest of the minority shareholders (La Porta et al., 1999; Burkart
and Panunzi, 2001). In this case the expropriation which minority shareholders have
to face and more generally, the other stakeholders, may create perverse effects such
as a reduction of the managers‟ incentives to invest and take initiatives (Burkart et
al., 1997). In short, one can conclude that although the concentration of ownership in
the hands of a large shareholder reduces the traditional agency problem between
managers and shareholders, it originates at the same time a new agency problem
between small and large shareholders.


      The board of directors, as the main internal mechanism for the solving of
agency conflicts within the firm (Jensen, 1993) centres its efforts on the supervision
of the managerial discretion. Empirical literature on this question seems to indicate
that the effectiveness of the board in its supervisory role depends on factors such as
its degree of independence, its size, the existence of specific committees or the
confluence in a single person (unitary leadership structure) of the responsibilities of
the CEO and the chairman of the board.
      Traditionally the degree of independence of board was linked to the presence
of outside directors – not belonging to the management team – as a way to guarantee
a better supervision of the management (Weisbach, 1988; Baysinger and Butler,
1985; Pierce and Zahra, 1992). On the contrary, other authors consider that the
participation of internal members favours the effectiveness of the board of directors
(Ocasio, 1994; Bhagat and Black, 1998). In this case the greater knowledge and
experience of the internal directors could improve the profitability of the firm.
      As for the size of the board, although the initial empirical evidence seemed to
suggest a negative relation between board size and the value of the firm (Yermack,
1996; Eisenberg et al., 1998), the latest research has begun to question this relation
(Dalton et al., 1999). The reasons justifying small boards –greater participation,
interaction and debate amongst members, greater problems of co-ordination and
communication in large boards (Lipton and Lorsch, 1992)– contrast with the
                                        -8-


advantages associated with large boards. A board with many members is more likely
than small one to generate alternative political coalitions that challenge the CEO and
take control over the firm. A large board also limits the possibility of the CEO
exerting social influence to maintain his power (Ocasio, 1994). The question
therefore remains open.
      The structuring of the board in different committees – compensation, audit or
finance being the most usual- seems to give the company more value (Klein, 1998).
The justification is that the committees facilitate the evaluation and supervision of
the acts carried out by the firm‟s management and they are more necessary as the
size of the firm increases.


      With regard to the structure of leadership, the theory literature favours the
posts of chief executive officer and chairman not being held by the same person
(Jensen, 1993), as a recommendation to guarantee a more effective supervision of the
actions of the CEO and his team of officers.


      Amongst the papers written on the boards of directors of financial firms, one of
the most important is that by Pi and Timme (1993). Using a sample of 112 North
American banks during the period 1987-1989 the authors show that in those boards
in which the posts of CEO and chairman are held by a single person, they are less
efficient at controlling processes of decision making by directors. This fact is
reflected in the lower profitability of the company, measured through the ROA ratio,
and lower cost efficiency, calculated using efficient frontier methodology. Simpson
and Gleason (1999) using a sample of 300 banks between 1989 and 1993 analyse
various aspects of the board, such as size, duality of the figure of CEO/chairman, and
the participation of the CEO and other directors in the firm‟s equity. The only
statistically significant result of the research shows the existence of a relation
between the unity of the leadership structure (CEO/chairman) and a reduction in the
probability of the bank going bankrupt. The justification makes reference to the fact
that a sole leadership provides a better strategic vision of the firm than an
independent chairman.
                                         -9-


      A pioneering work in Spain is that of Melle and Maroto (1999) who analyse
the influence of a specific group in the general assembly of Spanish savings banks.
Using a sample of 26 savings banks for the year 1997, they study what effects are
caused by the weight of local and regional governments in the assembly of savings
banks on their policies concerning loans and payment of employees. In the first case,
they find a positive relation between the percentage of governments in the assembly
and the loans granted to them. In the second case, the relation also proves to be
positive suggesting that both groups, employees and local and regional governments,
make a common front to pass by majority their mutual demands. Likewise, the study
investigates how the investment and financial policies in relation to the group of
local and regional governments affect the ROA ratio of the savings banks. This
relation proves to be negative so such so that savings banks with “abnormal” policies
for loan granting to the governments have a lower ROA.


      Continuing this same line of research, the paper presented here analyses the
relation existing between the composition and characteristics of the governing bodies
of Spanish savings banks and their level of efficiency.


2- SPANISH SAVINGS BANKS


2.1.- CHARACTERISTICS AND LEGAL FRAMEWORK
      Spanish savings banks are private financial foundations whose actions and
development allow us to classify them as financial intermediaries with their own
character within the Spanish Financial System. Specific features of this personality
are (Revell, 1991; García, 1994) (1) their work in foment saving, hence the
traditional activity of these savings banks until the last few years has been to attract
deposits; (2) their preferential attention to small customers and PYMES (small and
medium sized companies) through their extensive network of branches; (3) the
concentration of their business in a specific geographical area, the result amongst
other things of their origin as pawnshops, and the legal restrictions imposed on their
expansion by law until quite recent times. The main consequence of this fact has
                                           - 10 -


been the specialisation of savings banks in the economic, cultural and social aspects
of their region; and finally (4), the use of a part of their profits for cultural,
educational and social projects („community projects‟). In the last five years Spanish
savings banks have destined more than 3,700 million of euros to such areas.
      In short, the conclusion is that Spanish savings banks have been able to
combine their markedly social character with the maintenance of a financial activity
linked especially with the immediate region. A symbiosis which has enabled them to
become the undisputed leaders in the retail market.


      Together with these operational features, their organizational form is also
peculiar. Its most significant characteristic is that savings banks do not have alienable
title deeds. Thus, the structure and composition of their governing bodies is regulated
through laws whose origin is in the well-known LORCA, Law 31/1985 of 2 August
on the Regulation of basic norms of governance bodies of savings banks. This
consolidated The General Assembly, the Board of Directors and the Control
Committee as the principal governing bodies. In addition, this law implicates
different social groups in their management. These groups are related in one way or
another with the saving bank, its community or its area of action: local governments,
depositors, institutional founders and employees. This law also determined the exact
percentage of participation of each social group in the general assembly and their
maintenance in the other bodies.

                      PERCENTAGES GENERAL ASSEMBLY
                             STATE LAW 31/1985

                          11%         5%

                                                    44%        Depositors
                                                               Local governments
                                                               Founders
                       40%
                                                               Employees

      Figure 1. Source: the authors



      Thus, the general assembly, the same as general shareholders‟ meeting in
companies with shares, is principally responsible for endorsing the firm‟s annual
                                             - 11 -


accounts and electing the members who will form part of the board of directors and
the control committee. The board of director‟s primordial function is the financial
administration and management of the savings bank and the endorsement of the
budget destined to community projects. It is quite usual for the board to delegate
some of these functions to specific committees (amongst them the executive
committee has a special importance in charge of the day-to-day management of the
savings bank). The control committee is responsible for ensuring that the
management of the board of directors is done as efficiently and precisely as possible
within the lines set down by the general assembly. In all the governing bodies,
agreements are taken by simple majority of the members present except in very
specific cases. Finally, board of directors elects the chairman of the firm – to be in
charged of the official representation of the firm – and the CEO – to be responsible
for carrying out the agreements of the board-.


                    GOVERNING BODIES OF SPANISH
                          SAVINGS BANKS

                                GENERAL ASSEMBLY


                  BOARD OF                                      CONTROL
                  DIRECTORS                                    COMMITTEE



                      CEO
                  CHAIRMAN


   Figure 2. Source: Melle and Maroto (1999) and the authors



      Despite the fact that the LORCA applies to all Spanish savings banks, the
Constitutional Tribunal (in sentences 48 and 49/1988) allowed regional governments
to introduce modifications in virtue of the “eminently regional nature” of these
institutions. These modifications refer to the groups represented in the general
assembly –authorising the inclusion of “representatives of other social interests”–
and to their percentages of participation. As a result most regional governments have
                                               - 12 -


increased the percentages of local governments representation or introduced new
groups of interest in the representation of regional governments which has led to
even greater complexity in their make up (see figure 3).


      REGIONAL                                  PERCENTAGE IN THE ASSEMBLY
                                 No. S. Bks.
    GOVERNMENTS                                Founders   Employees Depositors Governments

  Andalucía                           6           9           7        28         56
  Asturias                            1          35           5        20         40
  Aragón                              2          10           7        41         42
  Baleares                            2          16           5        39         40
  Canarias                            2          10           5        26         44
  Cantabria                           1          10           5        22         63
  Castilla La Mancha                  2          10           7        22         61
  Castilla y León                     6           5          11        32         47
  Cataluña                           10         25-35       5-15      30-40      15-20
  Extremadura                         2          11           5        44         40
  Galicia                             2          25          10        40         25
  Madrid                              1          20           8        28         44
  Murcia                              1          15           7        30         48
  Navarra                             1          11           5        44         40
  La Rioja                            1          33           5        31         31
  C. Valenciana                       4           5          11        28         56
  País Vasco                          3          22           5        41         32

 Figure 3. Source: the authors




2.2.- THE PROBLEM OF CORPORATE GOVERNANCE IN SPANISH SAVINGS
BANKS
      The legal framework we have described above, characterised by the non
existence of alienable property rights, determines the special relevance in savings
banks of their internal control mechanism both to efficiently supervise the
managerial behaviour and to solve the organizational problem associated with its non
alienability. However, the same legal framework also determines that the control
rights in savings banks are assigned through laws to different social groups. This
situation explains the make up of a governance structure which generates problems
of asymmetric information and conflicts of interest since local and regional
governments, depositors, employees, founders and institutions of general interest
come together in the control of the savings bank, pursuing theoretically diverging
                                         - 13 -


interests. In this situation control rights are not assigned equally amongst all the
groups. Regional governments alter power relations in their favour, by
simultaneously assuming the double role of regulator and participant in the
governance of the savings banks. This dual function encourages the opportunist
behaviour of governments. Then they practically self assign the majority of control
rights on governing bodies. This as especially relevant fact for the board of directors
and the control committee as mechanisms responsible for the efficient management
of the savings bank and its supervision respectively. The lack of definition de iure of
the property rights in savings banks encourages regional governments to appropriate
them de facto through the control rights. The literature on the governance of the firm
refers to the potential expropriation by which minority shareholders are subjected to
the governing bodies of companies with shares, where the control is exercised by
large shareholders. In our case the directors of savings banks will act in favour of the
dominant group in the board of directors and in detriment to the rights of the rest of
the participants, especially depositors as the principal providers of funds in financial
firms. The result is that public authorities have a direct and significant influence on
investment and financing decisions of savings banks, and on the funds which are
earmarked for community projects. It is thus plausible to assume that savings banks
can adopt some operative and strategic decisions which are not guided so much by
criteria of economic efficiency as by political criteria. These decisions in the end
benefit the group of local and regional governments and have their reflection in a less
efficient bank. Obtaining greater financing (by buying debt securities issued by the
regional or local governments (Melle and Maroto, 1999)) or politically strategic
investment in local or regional projects are possible examples of what we could call
perquisites of governments.
      Consequently, the agency problem in Spanish savings banks, analogously to
the definition of Shleifer and Vishny (1997), refers to how the different stakeholders,
and especially the depositors, can ensure that the resources they contribute are not
wasted by local and regional governments.


      In this situation the traditional mechanisms of corporate governance seem
inoperative. Depositors in savings banks are generally small agents, fundamentally
                                               - 14 -


retail customers (free rider problem), uninformed (informative asymmetries),
dispersed and with diverging interests, who do not find sufficient incentives to
supervise the activities of the savings banks (Freixas and Rochet, 1999). To this lack
of incentives one must add other factors such as the existence of a peculiar system of
elections1 which make it particularly difficult for independent depositors to gain
access to governing bodies of savings banks, the non-remunerative nature of their
members or the existence of a deposit insurance which assures a sum of 20,000 euros
per depositor in case of bankruptcy. At the same time the supervisory aim for which
control committee was conceived is weakened from the moment in which its
structure and composition practically imitates that of the rest of the governing bodies
the general assembly and the board of directors. It thus seems likely that the control
committee will ratify decisions endorsed by the board of directors.
       As for the traditionally effective external mechanisms in financial companies
(capital market, market for corporate control, products and services market and
banking regulation), these do not seem able to solve the agency problem posed. The
organizational form of savings banks prevents the transmission of property rights and
therefore, the possibility of disciplining the actions of officers and governments
either through the capital market or the market for corporate control.
       What is more, the financial products market is characterised by offering a very
slight differentiation in the quality/price ratio, and a generalised policy of charging
commission for financial services (Melle and Maroto, 1999). To this we must add the
importance acquired by the CECA (Spanish Confederation of Savings Banks) in the
provision of certain services and products to the smallest savings banks which,
otherwise, these could not offer to their customers –investment funds, foreign
currency services, trading in futures, options etc.
       Thus, the regulation and supervision of the Bank of Spain is the only
mechanism to discipline the actions of the board. It has for years been requesting the


1
  To form part of the general assembly a depositor must be elected as a convention delegate from
amongst all the depositors of the savings bank by public draw held before a notary. In a second phase,
those delegates who wish to be elected as „consejeros delegados‟ representing the depositors must
present a candidacy backed by a number of depositors or delegates which varies according to the
statutes of each savings bank. This is a key aspect since some savings banks demand the signatures of
up to 1,500 depositors (e.g. Caja Duero) which in effect means that it is only consumer associations,
                                             - 15 -


state government to modify the LORCA and adapt the governing bodies of the
savings banks to the Cadbury report or similary to make the sector become more
professional. However, this possible reform needs to come from within the savings
banks themselves, since state government is not disposed to modify the legal
framework of the savings banks. As the future Financial Law is pending, up until
now the state government has limited itself to recommending caution in the
application of the LORCA so as not to distort the true objective for which the law
was passed. This objective was the democratisation of the governing bodies of
savings banks through involvement in their management of different groups
belonging to the geographical area in which the savings banks operate.


      The thesis we put forward in this research paper is therefore the following: a
greater participation of local and regional governments on the board of directors of
savings banks negatively conditions their degree of efficiency. This thesis is
specified in one basic hypothesis and three additional ones:


      The board of directors is directly responsible for approving the bank‟s
investment programme. Politicians have an important number of representatives
named from among local and regional governments because control rights of savings
banks are assigned by law. These representatives form a common front to defend
investment policies which encourage specific actions that benefit their interests.
Amongst others, the “concealed” obligation to participate in projects of the region
where it is situated, the buying of government stock issued by these authorities,
investment in local businesses, obtaining of loans at low interest, etc. Obviously
these investments are guided more by political motivation than economic ones and as
a result they can deteriorate the efficiency of savings banks.
H1: a greater percentage of votes of local and regional governments in the board of
directors negatively conditions cost efficiency of savings banks.




foundations or even political parties which are capable of placing their deposit candidates on the
general assembly.
                                         - 16 -


      The literature on corporate governance has considered the size of the board of
directors to be a key variable in monitoring management. Considering this, and
though the most recent empirical evidence is not conclusive, we presume that an
excessive number of directors will hinder the process of decision taking and,
consequently, that the size of the board negatively affects the efficiency of the
savings bank.
H2: a larger board of directors negatively affects cost efficiency of savings banks.


      The fundamental objective of the control commission is to control the board of
directors. However, the law assigns to the group of local and regional governments
what is practically a majority of the rights to vote in the commission. The
consequence is that the dual mechanism of governance is distorted since the same
group is dominant in both bodies. The control commission, therefore, will not
censure the decisions of the board of directors, and this dysfunction will negatively
affect the efficiency of the savings bank.
H3: a greater percentage of votes of local and regional governments in the control
committee has a negative influence on cost efficiency of savings banks.


      The board of directors of the savings banks has amongst its missions to
approve the capital budget that will be destined for the financing of community
projects. It seems logical to think, following our previous comments, that the
representatives of the local and regional governments tend to destine the biggest
possible quantity of financial resources to the funding of community projects.
Educational centres, promotion of sport, services for senior citizens, etc.. are some
of the activities in which savings banks are working.
H4: a greater percentage of votes of local and regional governments in the board has
a positive influence on the volume of funds destined for community projects.
                                            - 17 -


3- METHODOLOGY OF THE EMPIRICAL INVESTIGATION
3.1- SAMPLE AND SOURCES OF INFORMATION
The mergers undergone by some savings banks during the period of the study (1998
and 1999)2 and the impossibility of getting information as to the governing bodies in
others3 (some statutes have already been abolished) have reduced the study sample to
40 Spanish savings out of a total number of 49 (81% of the population for the year
1999).
      Financial data were obtained from the Statistics Yearbooks given by the
Confederation of Spanish Savings Banks and from the annual reports of each savings
bank. Information concerning the governing bodies was obtained from the social
statutes requested, firstly from the savings banks themselves and, after then, from the
mercantile register of each region.
3.2- VARIABLES
Dependant variables

EFIi: Cost efficiency of the savings bank for the year 1999, calculated using the
efficient frontier model proposed by Pastor (1996). Specifically this is a cost function
estimated with a determinist non parametric method called DEA (Data Envelopment
Analysis) used widely in many studies as a dependent variable representing the
efficiency of financial firms (Färe et al.,1985; Ferrier and Lovell, 1990; Grifell and
Lovell, 1994; Maudos et al., 2000;). These functions are determined through the
minimum cost level at which it is possible to produce a specific vector of outputs
given the prices of the inputs. Therefore for a set of observations the efficient frontier
in costs obtained is defined as the limit below which it is not possible to find any
observation. In our case the outputs and inputs used for their calculation were the
following: outputs: (1) sum of cash on hand and on deposits at central banks,
Government debt securities, due from credit institutions and investment securities;
(2) total loans and credits; (3) total deposits; (4) number of branches. Inputs: (a)
number of employees; (b) fixed assets (tangible). The price of input work is obtained
as the quotient between personnel expenses and number of employees, and that of


2
 Caja de ahorros de Vigo, Ourense y Pontevedra, Caja Municipal de Pamplona and Caja de ahorros
de Navarra.
                                             - 18 -


the capital factor from the quotient between administrative expenses and depreciation
of fixed assets (tangible).
       Data correspond to the balance sheets and statements of income for all the
Spanish savings banks for the year 1999. The efficient frontier is formed by those
savings banks whose efficiency value is one. These banks establish the reference for
the rest of the entities, whose efficiency is compared and measured in relation to the
frontier created. Thus, lower indexes of efficiency correspond to less efficient banks
(see figure 4).

EFI2i: Cost efficiency of the savings bank in the year 1999, calculated as operating
expenses to ordinary income ratio. To make our results robust we opted for use this
ratio as an alternative to the measurement of efficiency based on the calculation of
efficient frontiers. This ratio, commonly calculated by financial firms in their annual
reports, reflects the incidence of operating expenses on the results of the savings
bank, such that high values imply lower cost efficiency, and viceversa (see figure 4).

CPi: Community projects of savings banks in the year 1999 expressed as the quotient
funds destined to community projects divided by the savings bank and income before
tax.

Independent variables

ADCONSi: Participation of local and regional governments in the board of directors
of the savings bank during the years 1998 and 1999 (the same participation is
maintained), measured as the quotient between the number of members representing
local and regional governments and the total number of board members.

NMCONSi: Number of members on the board of directors during the years 1998 and
1999 (the same number is maintained).

ADCOM i: Participation of local and regional governments in the control committee
of the savings bank during the years 1998 and 1999 (the same participation is
maintained), measured as the quotient between the number of members representing
local and regional governments and the total number of control committee members.

3
 Caja de ahorros de Granada, Caja de ahorros de Jaén, Caja San Fernando de Sevilla y Jerez, Caja
Segovia and Caja de ahorros de la Inmaculada
                                           - 19 -




               EFFICIENCY OF SPANISH SAVINGS BANKS -YEAR 1999-
                                                   EFI       EFI2
    1   C.A. del MEDITERRÁNEO                       0.808    0.513
    2   C.A. y M.P. de ÁVILA                        0.791    0.465
    3   M.P. y C. General de A. de BADAJOZ          0.793    0.579
    4   C.E. i Pensions de BARCELONA - La Caixa     1.000    0.686
    5   C.E. de CATALUNYA                           1.000    0.675
    6   BILBAO BIZKAIA KUTXA                        0.941    0.430
    7   C.A. y M.P. del C.C.O. de BURGOS            0.951    0.534
    8   C.A. Municipal de BURGOS                    1.000    0.416
    9   C.A. y M.P. de EXTREMADURA                  0.952    0.587
   10   C.A. y Préstamos de CARLET                  0.774    0.794
   11   C.A. y M.P. de CÓRDOBA. CajaSur             0.697    0.555
   12   C.A. de GALICIA                             1.000    0.475
   13   C.A. de CASTILLA LA MANCHA                  0.753    0.660
   14   C.E. de GIRONA                              1.000    0.490
   15   C.A. Provincial de GUADALAJARA              0.861    0.714
   16   CAJA ESPAÑA                                 0.770    0.570
   17   C.A. de LA RIOJA                            0.911    0.539
   18   C.A. y M.P. de MADRID                       1.000    0.457
   19   UNICAJA                                     0.726    0.499
   20   C.E. Comarcal de MANLLEU                    0.749    0.585
   21   C.E. de MANRESA                             0.977    0.464
   22   C.E. LAIETANA                               0.738    0.491
   23   C.A. de MURCIA                              0.898    0.447
   24   C.A. y M.P. de ONTINYENT                    0.715    0.678
   25   C.A. de ASTURIAS                            0.746    0.484
   26   C.A. y M.P. de las BALEARES                 0.619    0.634
   27   C. Insular de A. de CANARIAS                0.657    0.704
   28   C.A. de POLLENÇA                            1.000    0.638
   29   C.E. de SABADELL                            0.781    0.578
   30   C.A. de SALAMANCA y SORIA                   1.000    0.543
   31   C.A. GIPÚZKOA y SAN SEBASTIÁN               0.794    0.460
   32   C. General de A. de CANARIAS                0.903    0.515
   33   C.A. de SANTANDER Y CANTABRIA               0.634    0.687
   34   M.P. y C.A. de HUELVA y SEVILLA             0.698    0.565
   35   C.E. de TARRAGONA                           1.000    0.558
   36   C.E. de TERRASSA                            0.894    0.567
   37   BANCAJA                                     1.000    0.536
   38   C.E. del PENEDÉS                            0.891    0.523
   39   C.A. de VITORIA y ÁLAVA                     0.863    0.481
   40   IBERCAJA                                    1.000    0.566


Figure 4.Source: the authors
                                                 - 20 -




Control variables: (1) Variable of debt of savings banks calculated as the quotient
between total liabilities and savings bank‟s equity (LEVERAGE) (Arthur, 2001). A
priori a direct relation with efficiency is expected, so the higher the debt ratio, the
external resources of the entity are greater and the free cash flow available to
management is lower. For the same reason, an negative relation is expected with the
sum of funds destined to community projects. (2) Variable of average size of
branches calculated as total deposits to number of branches ratio (DEPOFIC) (Azofra
and de La Fuente, 1989; Brook et alter, 2000). A priori a positive relation is
expected. Higher ratio values indicate better use of the bank‟s fixed structure and
therefore greater levels of efficiency; (3) variable of financial innovation of the
savings bank calculated as the quotient between total operations in memorandum
accounts4 and the total assets of the savings bank (INNOVA). With this variable we
try to characterise what degree of the business of savings bank is not traditional. A
priori, the sense of the relation cannot be determined (Berger and Mester, 1997). (4)
An instrumental variable for community projects of savings banks in the previous
year (1998) calculated as funds destined to community projects divided by income
before taxes (CP98). A priori a positive relationship is expected with the CP variable.
It is believed that the reason for this is that a large part of the community projects of
savings banks for a designated year is a direct consequence of the expenses of social
projects form the previous year. (5) Variables of size. On one hand we use the natural
logarithm of the average total assets of the bank (LNATM), and alternatively, three
dummy variables that indicate the size at different stages: ACT1 that takes the value
one for banks with average total assets between 2,100 m. and 3,300 m. of Euros, and
zero in the opposite case. ACT2 that takes the value one for banks with average total
assets between 3,300 m. and 7,800 m. of Euros, and zero in the opposite case. ACT3
that takes the value one for banks with average total assets greater than 7,800 m. and
zero in the opposite case. (Berger and Mester, 1997; Berger and Hannan, 1998).
Values used for each dummy variable were calculated based on quartiles. One of the

4
  Similar to off balance sheet in American banks. These accounts include financial futures on
securities and interest rates, options, outstanding currency and financial assets sales and purchases and
other interest-rate transactions.
                                         - 21 -


advantages of using these variables is that they permit one to test the existence of non
monotonous and non linear relations between the efficiency and the size of the
savings bank. The sense of these relations cannot be determined a priori.

3.3- DESCRIPTIVE ANALYSIS
      The descriptive analysis of variables appears in detail in the table below and
includes the average value, standard deviation and maximum and minimum values of
each variable. In the annex there is also a matrix of correlations of the independent
variables.

                            DESCRIPTIVE ANALYSIS

          N obs. Minimum Maximun Median Average Std. Dev.
  EFI      40        0.619   1.000  0.877  0.857     0.123
  EFI2     40        0.416   0.795  0.549  0.559     0.089
  CP       40        0.067   0.919  0.187  0.233     0.157
  ADCONS   40        0.167   0.750  0.471  0.440     0.164
  ADCOM    40        0.143   0.714  0.400  0.396     0.143
  NMCONS   40       10.000  21.000 17.000 16.375     2.724
  CP98     40        0.099   0.847  0.170  0.206     0.122
  LEVERAGE 40        8.100   20.00 13.495 13.225     2.513
  DEPOFIC  40        7.051   8.879  7.608  7.689     0.372
  INNOVA   40        0.079   0.292  0.177  0.171     0.051
  LNATM    40        9.893  16.137 13.260 13.297     1.305

             Cost efficiency measured through the frontier functions has an average
value of 0.857 and a standard deviation of 0.123. These values nearly correspond
with the results obtained by Pastor (1996) in his study on the efficiency of savings
banks. The average value is indicative that there is still potential room for
improvement of savings banks of fourteen percentage points.
      As for the cost efficiency calculated through operating expenses to ordinary
income ratio, the range of values obtained is between 0.42 –maximum efficiency
level corresponding to Caja de Burgos- and 0.79 –minimum efficiency level
corresponding to Caja de ahorros y préstamos de Carlet-.


      Savings banks destine 23% of their income before tax to community projects.
The minimum value corresponds to Caixa del Penedés (with 7%) and the maximum
                                         - 22 -


to Caja Insular de Canarias (with 91%). We decided to exclude from the study of
community projects this latter value as we considered it extreme, since its deviation
with respect to the average is four times higher than the standard deviation. In
addition, the second highest value on the scale is 65% very far from the extreme
value excluded.


      The average number of votes that the local and regional governments have on
the board of directors of the savings banks is around 44% of the total. Boards of
directors with a lower political representation (17%) correspond to savings banks
situated in Cataluña. The savings bank with the greatest political influence (75%) is
Caja Guadalajara. We obtained similar results for the control committee. The average
is situated a little lower (39%), and the maximum and minimum values coincide
more or less with those of the board (14% and 71%). The savings banks with these
values are, starting with the lowest, the nine savings banks of Cataluña, and at the
upper limit, Caja Castilla La Mancha.


      The average number of members who participate on the board of directors of
savings banks is around 16, oscillating between the 10 of Caja de Sabadell and the 21
of La Caixa, Caixa de Catalunya, Bancaja and Caixa Galicia. This average value is
certainly high in comparison with two other studies carried out for non financial
firms. However, we must remember that Spanish savings banks have limited by law
the number of members who may participate in the board of directors. Observing the
matrix of correlations we see, in agreement with other empirical studies, the
correlation existing between the size of the savings bank (LNATM) and the number
of members on the board of directors. This result is in agreement with the fact that
when the savings bank is big, its geographical radius of action is greater and hence,
the number of depositors, politicians and institutions of general interest in each
region which it will have to integrate in its governing bodies is higher. In these cases,
to achieve a more efficient management, savings banks normally appoint executive
committees in which only a part of the members of the board of directors take part,
maintaining the proportions of representation of each group. Unfortunately we were
not able to obtain these data for the study. Nevertheless, as a consequence of the high
                                         - 23 -


correlation between the variables NMCONS-LNATM and DEPOFIC-LNATM we
chose to use the dummy variables of size specified earlier.


3.4- METHODOLOGY
      The four hypotheses posed are tested using the estimation of three Tobit
equations. The basic reason for not using ordinary least square regressions is that the
dependent variables (EFI, EFI2 and CP) are censored between the values zero and
one. For this type of variable the most suitable treatment is with Tobit models.
      The regression equation that we formulate for the empirical verification of the
first and second hypotheses referring to the composition of board of directors is as
follows:

EFIi(2) = 0 + 1ADCONS + 2 NMCONS + 3 LEVERAGE + 4 DEPOFIC +
5 INNOVA + 6 ACT1 + 7 ACT2 + i                (1)


      The equation we pose for the empirical verification of the third hypothesis
referring to the participation of local and regional governments on control committee
is the following:

EFIi (2) = 0 + 1ADCOM + 2 LEVERAGE + 3 DEPOFIC + 4 INNOVA + 5
ACT1 + 6 ACT2 + 7 ACT3 + i           (2)


      And finally, the regression that we use for the empirical verification of the
fourth hypothesis referring to the share out for community projects is as follows:

CPi = 0 + 1 ADCONS + 2 CP98 + 3 LEVERAGE + 4 ACT1 + 5 ACT2 +
6 ACT3 + i           (3)


3.5.- RESULTS
First equation: hypotheses 1 and 2 (composition of the board of directors)

      The results obtained (table I) in the estimation of the equation with the
dependent variable EFI (efficient frontiers) show statistically significant coefficients
for the variables ADCONS, NMCONS and LEVERAGE. Thus, firstly there is
                                                         - 24 -


evidence of the existence of an inverse relation between the participation of local and
regional governments on the board and the variable EFI. This seems to confirm the
first of our hypotheses, that is, a greater presence of politicians in savings banks has a
negative influence on their efficiency. Secondly, the relation between cost efficiency
and the number of members on the board is a positive relation, contrary to what is
expected. The justification may lie in the fact that the savings banks which have a
large number of members on their boards –which are usually those of greater size-
tend to create executive commissions (with a notably lower number of members) in
order to more efficiently manage their activities. Therefore, the existence of
executive commissions in the largest savings banks would mean their greater
efficiency such that the relation between size and efficiency would become direct.

                                 Table I - Hypotheses 1 and 2

                    COMPOSITION OF BOARD OF
             DIRECTORS AND COST EFFICIENCY (EFI y EFI2)

      DEPENDENT V.                                         EFI                        EFI2

      CONSTANT                                              0.0549                      1.4396 ***
                                                          (0.5547)                    (0.2831)
      ADCONS                                               -0.3206 **                   0.1827 **
                                                          (0.1583)                    (0.0771)
      NMCONS                                                0.0312 ***                 -0.0037
                                                          (0.0106)                    (0.0048)
      LEVERAGE                                              0.0219 **                   0.0014
                                                          (0.0099)                    (0.0048)
      INNOVA                                                0.6905                      0.0047
                                                          (0.4935)                    (0.2482)
      DEPOFIC                                               0.0090                     -0.1195 ***
                                                          (0.6899)                    (0.0358)
      ACT1                                                  0.0166                      0.0135
                                                          (0.0575)                    (0.0296)
      ACT2                                                 -0.0763                     -0.0126
                                                          (0.0567)                    (0.0294)

      -2Log likelihood                                       9.8709                     48.8237
      Prob>Chi2                                               0.016                       0.017
                  In brackets the value of the standard error of the statistical coefficient.
          ***, **, *, indicate variables statistically significant at 99%, 95% y 90% respectively.
                                         - 25 -


      Finally, the results of this first equation prove a direct relation, just as we
expected a priori, between the control variable of debt and efficiency.
      In the equation with the dependent variable EFI2 (ratio), the results again
confirm the existence of a statistically significant relation between efficiency and the
participation of local and regional governments on the board of directors. In this
case, by inverting the sense of the dependent variable (greater values of EFI2
indicate smaller levels of efficiency) the relations that are obtained are also inverse
such that the sign of the variable ADCONS turns out to be positive. Thus, greater
political participation on the board corresponds to a greater value of the ratio
operating expenses to ordinary income and, hence, to a smaller degree of efficiency.
What is more, the results of this second estimation do not show, on the contrary to
what happened in the first equation, a statistically significant relation between the
size of the board and efficiency, although the direct sense of the relation is
maintained. As for the control variables, only DEPOFIC turns out to be significant
with a negative sign, which implies that a better usage of the fixed structure of the
company positively influences its cost efficiency.


      In conclusion, the results allow us to corroborate the basic hypothesis on the
participation of local and regional governments on the board, but not the hypothesis
referring to the size of the board.


Second equation: hypothesis 3 (composition of the control committee)

      The results obtained in the estimation of the second equation are shown in table
II (below) for the two dependent variables. Thus, while in the regression with the
dependent variable EFI2 the participation of local and regional governments on the
control committee is shown to be significant at 95%, in the equation with the variable
EFI this participation does not prove to be statistically significant (although the sign
of the relation is the predicted sign). Because of this, we cannot conclude that the
percentage of public authorities on the control committee affects, either positively or
negatively, the efficiency of the savings banks. As for the control variables, as in the
equation on the board of directors, the coefficients of the variable LEVERAGE and
                                                         - 26 -


DEPOFIC prove to be statistically significant. The results also show a significant and
positive relation between the variable ACT3 (banks with average total assets above
7,800 m. of Euros and cost efficiency calculated by frontier functions. Although this
relation seems to indicate to priori that the larger savings banks are the most
efficient, this is not corroborated with the dependent variable EFI2.


                                      Table II - Hypothesis 3

                PARTICIPATION OF LOCAL AND REGIONAL
                GOVERNMENTS ON CONTROL COMMITTEE
                   AND COST EFFICIENCY (EFI y EFI2)

      DEPENDENT V.                                         EFI                        EFI2

      CONSTANT                                              0.4940                      1.4548 ***
                                                          (0.6145)                    (0.3099)
      ADCOM                                                -0.2070                      0.1875 **
                                                          (0.1753)                    (0.0889)
      LEVERAGE                                              0.0206 *                    0.0010
                                                          (0.0102)                    (0.0049)
      INNOVA                                                0.4691                      0.0002
                                                          (0.5270)                    (0.2635)
      DEPOFIC                                               0.0080                     -0.1274 ***
                                                          (0.0761)                    (0.0384)
      ACT1                                                  0.0552                      0.0106
                                                          (0.0683)                    (0.0345)
      ACT2                                                  0.0158                     -0.0205
                                                          (0.0721)                    (0.0367)
      ACT3                                                  0.1598 **                  -0.0082
                                                          (0.0776)                    (0.0374)


      -2Log likelihood                                         7.617                     48.343
      Prob>Chi2                                                0.081                      0.024
                  In brackets the value of the standard error of the statistical coefficient.
          ***, **, *, indicate variables statistically significant at 99%, 95% y 90% respectively.




Third equation: hypothesis 4 (resources destined to community projects)

   The test of the fourth hypothesis gives the values shown in table III (below). The
results show that the coefficient of the variable ADCONS is statistically significant
                                                         - 27 -


(at 95%) with a positive sign. These results allows us to affirm, with some caution,
that the presence of politicians on the board of directors positively conditions the
percentage of income that the savings bank destines to community projects.



                                      Table III- Hypothesis 4

             PARTICIPATION OF LOCAL AND REGIONAL
             GOVERNMENTS ON BOARD OF DIRECTORS
           AND EXPENDITURE ON 'COMMUNITY PROJECTS'

      DEPENDENT V.                                                      OBS

      CONSTANT                                                    -0.0317
                                                                 (0.1214)
      ADCONS                                                       0.1817 **
                                                                 (0.0829)
      CP98                                                         1.0251 ***
                                                                 (0.2369)
      LEVERAGE                                                    -0.0008
                                                                 (0.0061)
      ACT1                                                          -0.274
                                                                 (0.0380)
      ACT2                                                        -0.0438
                                                                 (0.0385)
      ACT3                                                        -0.0253
                                                                 (0.0393)

      Log likelihood                                                36.649
      Prob>Chi2                                                      0.000
                  In brackets the value of the standard error of the statistical coefficient.
          ***, **, *, indicate variables statistically significant at 99%, 95% y 90% respectively.




On the other hand the coefficients for the control variable is statistically significant
for the community projects of the year 1998 (this is significant at 99% as a positive
indicator just as we had predicted).
                                         - 28 -


4- CONCLUSIONS

      The paper presented here analyses the effect of the participation of local and
regional governments in the governing bodies of Spanish savings on cost efficiency
and the volume of resources destined to community projects during the period 1998-
1999. Using a sample of 40 savings banks the results of the study show that, although
there are many factors which influence the efficiency of these banks, the
participation of local and regional governments in the board of directors must be
included amongst them. Thus, we find existence of a negative and significant relation
between the participation of politicians on the board of directors of savings banks
and their cost efficiency in costs. In addition, we find a positive and significant
relation between political participation in the board and the channelling of more
financial resources towards community projects.
      Of course we are aware that it is necessary to take these results with some
caution, especially as we are dealing with a cross-sectional study. However, this does
not change the reality of the results obtained. This results seem to confirm the
perverse effect of political participation on the board on the efficiency of savings
banks. Our results permit us to offer empirical evidence which favours the need to
establish limits to the intervention of public authorities in the governance of savings
banks. Such restrictions would be aimed at re-establishing the balance of powers
within the governing bodies as a way to guarantee that the decisions adopted be
based on criteria of professionality and competition and not on political criteria. This
limitation would not be incompatible with the assignation of more financial resources
to community projects. The direct relation confirmed between participation of local
and regional governments on the board of directors and the funds spent on
community projects cannot be used as a justification for the greater presence of
politicians on the board. First, because it is not a matter of increasing the efficiency
and reducing the community projects of the savings bank, as some seem to think, but
to improve the efficiency and thus obtain higher profits so as to be able to take on
more community projects. So efficiency and community projects are inseparable.
And secondly, because it would be a mistake to identify social aims with political
goals. Savings bank are not government companies destined to provide citizens with
                                          - 29 -


public services, but private companies – or private foundations if one prefers – which
dedicate a part of their profits to investments of a social nature.
      Our study reveals in addition that though there is a negative relation, but
statistically insignificant, between the variables participation of politicians on the
control committee and efficiency, the fact is that the savings banks are less efficient
with a greater number of politicians on the committee. We can therefore make a
conjecture as to the incapacity of the control committee to perform the mission
assigned it by law, which is that of controlling the board.


      Finally, together with the results we have presented, it is observed that the
relation between the number of members on the board and the efficiency of savings
banks cannot be considered proven. Despite the fact that the result is statistically
significant using as variable the efficiency calculated by the efficient frontiers
method, this result is not maintained when the variable is replaced by the ratio
operating expenses to ordinary income.


      In summary, we believe that our work contributes to a further clarification of
the study of corporate governance of savings banks. Future lines of investigation
would need to broaden the time scope of the work and complete the information
referring to the governing bodies of savings banks (executive committees, number of
board meetings, where exactly each board member comes from -political parties,
consumer associations, …-).
                                                 ANNEX




                                                 MATRIX OF CORRELATIONS

           ADCONS    ADCOM    NMCONS    CP98       LEVERAGE   INNOVA    DEPOFIC   LNATM    ACT1     ACT2     ACT3
ADCONS      1.000
ADCOM       0.845     1.000
NMCONS      0.328     0.310    1.000
CP98        0.091    0.0961    0.237    1.000
LEVERAGE    -0.099   -0.082    -0.130   0.033        1.000
INNOVA      0.304     0.307    0.223    0.040        -0.178    1.000
DEPOFIC     0.221     0.285    0.315    0.036        -0.312    0.148      1.000
LNATM       -0.117    0.021    0.636    0.185        -0.218    0.277      0.445   1.000
ACT1        -0.069   -0.017    -0.252   -0.143       0.055     -0.165     0.037   -0.116   1.000
ACT2        0.105     0.166    0.112    -0.095       -0.119    -0.049     0.259   0.156    -0.333   1.000
ACT3        -0.004    0.036    0.563    0.211        -0.065    0.371      0.191   0.684    -0.333   -0.333   1.000
BIBLIOGRAPHY
Akella, S. R. y Greenbaum, S. I., 1988, “Savings and loan ownership structure and
    expense-preference”, Journal of Banking and Finance, 12, págs. 419-437.
Allen, F. y Cebenoyan, A. S., 1991, “Bank acquisitions and ownership structure:
    theory and evidence”, Journal of Banking and Finance, 15, págs. 425-448.
Anderson, R.C. y Fraser, D.R., 2000, “Corporate control, bank risk taking, and the
    health of the banking industry”, Journal of Banking and Finance, 24, págs.
    1383-1398.
Arthur, N., 2001, “Board composition as the outcome of an internal bargaining
    process: empirical evidence”, Journal of Corporate Finance, 7, págs. 307-340.
Azofra, V., 2000, “El problema del gobierno de la empresa”, proyecto de
    investigación de acceso a plaza del cuerpo de catedráticos de universidad.
Azofra, V. y de la Fuente, J., 1989, “La racionalidad económica de las fusiones
    bancarias: el caso de las cajas de ahorro castellano-leonesas”, Anales de Estudios
    Económicos y Empresariales, 4, págs. 163-181.
Baysinger, B. D. y Butler, H. N., 1985, “Corporate governace and the board of
    directors: performance effects of changes in board composition”, Journal of
    Law, Economics and Organization, 1, págs. 101-124.
Berger, A. N. y Mester, L. J., 1997, “Inside the black box: what explains differences
    in the efficiencies of financial institutions?”, Journal of Banking and Finance,
    21, págs. 895-947.
Berger, A. N. y Hannan, T. H., 1998, “The efficiency cost of market power in the
    banking industry: a test of the quiet life and related hypotheses”, The Review of
    Economics and Statistics, 80, págs. 454-465.
Bhagat, S. y Black, B., 1998, “The relationship between board composition and firm
    performance”, in Comparative Corporate Governance edited by K.J. Hopt et al.
Brickley, J.A. y James, C.M., 1987, “The takeover market, corporate board
    composition, and ownership structure: the case of banking”, Journal of Law and
    Economics, 30, págs. 161-181.
Brook, Y., Hendershott, R. J. y Lee, D., 2000, “Corporate governance and recent
    consolidation in the banking industry”, Journal of Corporate Governance, 6,
    págs. 141-164.
                                        - 32 -


Burkart, M.; Gromb, D.; Panunzi, F., 1997, “Large shareholders, monitoring, and the
    value of the firm”, The Quarterly Journal of Economics, 112(3), págs. 693-728.
Burkart, M. y Panunzi, F., 2001, “Agency conflicts, ownership concentration and
    legal shareholder protection”, Discussion Paper Series No. 2708, Centre for
    Economic Policy Research.
Caruana, J., 2001, Discurso pronunciado en la Asamblea General de la CECA el 18
    de Abril de 2001.
Ciancanelly, P. y Reyes, J. A., 2000, “Corporate governance in banking: a conceptual
    framework”, Paper presentado en la European Financial Management
    Association Conference, Athens, Junio 2000.
Dalton, D. R.; Daily, C. M.; Johnson, J. L.; Ellstrand, A. E., 1999, “Number of
    directors and financial performance: a meta-analysis”, Academy of Management
    Journal, Diciembre.
Demsetz, H. y Lehn, K., 1985, “The structure of corporate ownership: causes and
    consequences”, Journal of Political Economy, 93(6), págs. 105-127.
Eisenberg, T.; Sundgren, S.; Wells, M., 1998, “Larger board size and decreasing firm
    value in small firms”, Journal of Financial Economics, 48, págs. 35-54.
Färe, R.; Grosskopf, S.; Logan, J., 1985, “The relative performance of publicy-
    owned and privately-owned electric utilities”, Journal of Public Economics,
    26(1), págs. 89-106.
Fernández, A. I., 2001, El gobierno de la empresa, mimeo Curso de Doctorado
    Nuevas Tendencias en Dirección de Empresas, Universidad de Burgos.
Ferrier, G. y Lovell, C. A. K., 1990, “Measuring cost efficiency in banking:
    Econometric and linear programing evidence”, Journal of Econometrics, 46,
    págs. 229-245.
Freixas, X., 1998, “Nuevos avances en la teoría de la regulación financiera”, Moneda
    y Crédito, 206, págs. 11-42.
Freixas, X. y Rochet, J. C., 1999, Economía Bancaria, Ed. BBV y Antoni Bosch.
García Roa, J., 1994, Las cajas de ahorros españolas: cambios recientes, fusiones y
    otras estrategias de dimensionamiento, Ed. Civitas.
Glassman, C.A. y Rhoades, S.A., 1980, “Owner vs. Manager control effects on bank
    performance” Review of Economics and Statistics, 62, págs. 263-270.
                                        - 33 -


Gorton, G, y Rosen, R., 1995, “Corporate control, portfolio choice, and the decline of
    banking”, The Journal of Finance, 5, págs. 1377-1420.
Grifell, E. y Lovell, C. A. K., 1994, “A generalized Malmquist productivity index”,
    Working Paper, University of Georgia.
Grossman, S.J. y Hart O.D., 1980, “Takeovers bids, the free-rider problem and the
    theory of the corporation”, The Bell Journal of Economics, 11, págs. 42-64.
Hannan, T. H. y Mavinga, F., 1980, “Expense preference and managerial control: the
    case of the banking firm”, Bell Journal of Economics, 11, págs. 671-682.
Hirschey, M., 1999, “Managerial equity ownership and bank performance:
    entrenchment or size effects?”, Economic Letters, 64, págs. 209-213.
Hubbard, R. G. y Palia, D., 1995, “Executive pay and performance: evidence from
    the U.S. banking industry”, Journal of Financial Economics, 39, págs. 105-130.
Jensen, M., 1993, “The modern industrial revolution, exit, and the failure of internal
    control systems”, Journal of Finance, 48(3), págs. 831-880.
Jensen, M. y Meckling, W., 1976, "Theory of the firm: Managerial bahavior, agency
    costs and ownership structure", Journal of Finance Economics, 48, págs 831-
    880.
Klein, A., 1998, “Firm performance and board committee structure”, Journal of Law
    and Economics, 41, págs. 275-303.
La Porta, R.; López de Silanes, F.; Shleifer, A., 1999, “Corporate ownership around
    the world”, Journal of Finance, 54(2), págs. 471-517.
Lipton, M. y Lorsch, J. W., 1992, “A modest proposal for improved corporate
    governance”, Business Lawyer, 48, págs. 59-77.
Maudos, J.; Pastor, J.M. y Serrano, L., 2000, “Convergence in OECD countries:
    Technical change, efficiency and productivity”, Applied Economics, 32(6), págs.
    757-765.
Melle, M. y Maroto, J.A., 1999, “Una aplicación del gobierno de empresas:
    incidencia de las administraciones públicas en las decisiones asignativas de las
    cajas de ahorro españolas”, Revista Europea de Dirección y Economía de la
    Empresa, 8(2), págs. 9-40.
                                        - 34 -


Morck, R., Shleifer A. y Vishny, R., 1988, “Management ownership and market
    valuation: an empirical analysis”, Journal of Financial Economics, 20, págs.
    293-316.
Ocasio, W., 1994, “Political dynamics and the circulation of power: CEO succession
    in U.S. industrial corporations, 1960-1990”, Administrative Science Quarterly,
    39, págs. 285-312.
Pastor, J. M., 1996, “Diferentes metodologías para el análisis de la eficiencia de los
    bancos y cajas de ahorros españolas”, Documento de Trabajo núm. 123,
    Fundación FIES.
Pearce, J. A. y Zahra, S. A., 1992, “Board compensation from a strategic contingency
    perspective”, Journal of Management Studies, 29, págs. 411-438.
Pi, L. y Timme, S.G., 1993, “Corporate control and bank efficiency”, Journal of
    Banking and Finance, 17, págs. 515-530.
Prowse, S., 1997a, “The corporate governance system in banking: what do we
    know?”, BNL Quarterly Review, Special issue, págs. 10-40.
Prowse, S., 1997b, “Corporate control in commercial banks”, The Journal of
    Financial Research, vol. XX, núm. 4, págs. 509-527.
Rajan, R. G. y Zingales, L., 1998, “The governance of the new enterprise”, Working
    Paper, University of Chicago.
Revell, J., 1991, "Consecuencias de los cambios recientes en las cajas de ahorros de
    europa occidental", Papeles de Economía Española, 46, págs 173-202.
Schranz, M.S., 1993, “Takeovers improve firm performance: evidence from the
    banking industry”, Journal of Political Economy, 101, págs. 299-326.
Simpson, W. G. y Gleason, A. E., 1999, “Board structure, ownership, and financial
    distress in banking firms”, International Review of Economics and Finance, 8,
    págs. 281-292.
Shleifer, A. y Vishny, R. W., 1997, " A survey of corporate governance", The
    Journal of Finance, 52(2), págs. 737-783.
Weisbach, M. S., 1988, “Outside directors and CEO turnover”, Journal of Financial
    Economics, 20, págs. 431-460.
Yermack, D., 1996, “Higher market valuation of companies with a small board of
    directors”, Journal of Financial Economics, 40, págs. 185-211.
- 35 -

								
To top