Documente Academic
Documente Profesional
Documente Cultură
MONITORING FIRMS
Elisabeth Paulet
List of figures ix
List of tables x
Preface xiii
Acknowledgements xiv
INTRODUCTION 1
vi
CONTENTS
3.1 Introduction 76
3.2 The value of association with the Crédit Mobilier on cash
flow, investment and market value 78
3.3 Transformation of the raw data according to ‘traditional
financial accounting’ 91
3.4 Extensions of the results: description and interpretation
of a robustness test 98
3.5 The expression of all variables in real terms 104
3.6 General conclusion 108
APPENDIX A
Data on the Crédit Mobilier 136
vii
CONTENTS
APPENDIX B
Data relative to the general indices (GNP, share prices)
for France between 1852 and 1914 140
B.1 General price indices 140
B.2 Share price series 141
APPENDIX C
Data relative to the affiliated companies 142
C.1 The whole data 142
C.2 The Compagnie Immobilière 142
APPENDIX D
Data relative to non-affiliated companies 146
APPENDIX E
Monthly share prices for affiliated companies
between 1866 and 1868 148
APPENDIX F
Data for investment and cash flow tests 152
Notes 161
Bibliography 166
Index 174
viii
FIGURES
ix
TABLES
x
TABLES
xi
TABLES
xii
PREFACE
xiii
ACKNOWLEDGEMENTS
To Patrick.
xiv
INTRODUCTION
• Why might banks want and be able to obtain the power to influence and
monitor managers?
• Why might shareholders benefit if the bank has such power?
Agency and signalling theory are concerned with the information and
interest conflicts which can exist between the different economic actors
associated with the firm: managers, shareholders and creditors. This
theory holds that some agents (the managers) are better informed about
the quality of the enterprise than the principals (shareholders, creditors).
To resolve the conflict of interest, contracts (implicit or explicit) must be
negotiated between the lenders, the creditors, the shareholders and the
managers.
The term ‘agency’ derives from the fact that corporate decisions are
delegated to agents (e.g. managers) who act on behalf of other parties
(external investors). The relation between investment and debt is explained
by the choice between debt and equity for the financing of a project.
The signalling approach makes the assumption that there is asymmetric
information as between management and investors: the managers have
more information than the shareholders on the value of their enterprise.
1
THE ROLE OF BANKS IN MONITORING FIRMS
First the role played by the Crédit Mobilier on the stock market is
determined. The level of volatility will be quantified in order to assess the
2
INTRODUCTION
French general economic context between 1852 and 1913. Share price series
are then used to investigate the impact of the bankruptcy of the Crédit
Mobilier on the stock market for affiliated and non-affiliated companies.
Share prices did not fall when the bank failed. The conclusion is that the
Crédit Mobilier provided liquidity to firms, but apparently did not seem to
be supervising them, and its seal of approval was probably not worth as
much as that of J.P.Morgan1 or the Großanken.2
Some remarks must be made concerning the data used to construct the
test. During the period statistical sources such as share prices and
balance sheets were not always kept according to strict rules. In addition
the amount of statistical information available over the period was quite
limited because managers were not obliged to publish their accounts. So
the validity of our results has to be assessed in the context of a panel
data of twenty five companies which provided information on their
financial situation.
As the Crédit Mobilier seems not to have succeeded in controlling
firms, it would be interesting to measure the effect of an association with
this establishment on investment policy. The basic theoretical idea is as
follows. In their world of perfect information and no agency costs, firms
can borrow and invest optimally; liquidity has no effect on the
investment decision. In their classic paper, Franco Modigliani and
Merton Miller (1958) showed under what conditions financial structure
becomes irrelevant for real investment decisions. In a context of
asymmetric information this proposition is no longer true. The next step
in our research is to analyse the relationship between investment and
cash flows between 1860 and 1880.
The effect on investment policy of an association with this
establishment will then be measured and compared to Großbanken
policy during a similar period. This proposed analysis then tests the
proposition that the Crédit Mobilier acted like a German-style universal
bank. To that intent it examines the effects on the investment behaviour
of French firms of affiliation with the Crédit Mobilier. Investment
equations for a modified version of Tobin’s Q-model are estimated.
Two samples are considered: the first consists of corporations that were
affiliated to the Crédit Mobilier, the second includes companies which
had to resort to other sources of finance. The focus of this construction
is on contrasting the financing of the Crédit Mobilier’s affiliated
enterprises with those which, for one reason or another, did not form
part of the nexus.
The chronological limits of this study are imposed by the data sources:
after 1880 the Crédit Mobilier lost its powerful role in the banking sector
and thus no records exist after this point.
The sample includes fifteen companies: seven were affiliated with the
Crédit Mobilier, eight with the Crédit Lyonnais. As each company, in any
3
THE ROLE OF BANKS IN MONITORING FIRMS
event, had a relationship with a bank, the interpretation of the results must
be in some sense cautious.
Using this evidence, the following two conclusions are reached:
1 companies that maintained close ties with the Crédit Mobilier were less
liquidity constrained after the bank’s crisis in 1867 than before;
2 unaffiliated companies experienced no more difficulty in financing
projects than those which resorted to the Crédit Mobilier.
• the bank prevents the firm from undertaking projects that are too risky;
• the firm signals its reputation to the market by having a regular relation
with a bank.
4
INTRODUCTION
5
1
AGENCY THEORY
AND MONITORING
A theoretical and empirical interpretation
1.1 INTRODUCTION
In this chapter the basic concepts of agency and signalling theories applied
to the financing of the firm are presented. In particular the chapter will
study how a debt-equity contract is defined between the manager of a firm,
the shareholders and the bank which provides liquidity.
Agency theory, although not quite new, is a part of decision theory
which is currently characterised by a dynamic development. The well-
known basic model of decision theory is related to a situation in which
one person—the so-called decision-maker—has to make a decision as well
as to bear its consequences. This assumption is often unrealistic.
Therefore, in agency theory the decision-maker (the agent) and the
‘beneficiary’ of the decision (the principal) are all distinct. A typical
example concerns the management of the firm: the agent, the manager of
the firm, chooses an investment project. The principal—the shareholders of
the firm and (or) the bank which finances the enterprise—partly support
the risks involved in this investment.
Hence, the common element is the presence of at least two individuals.
The first (the agent) must choose an action from a number of alternative
possibilities. The action affects the welfare of both the agent and the
principal. The principal has the additional function of prescribing payoff
rules; that is, before the agent chooses the action, the principal determines a
rule that specifies the fee to be paid to the agent as a function of the
7
THE ROLE OF BANKS IN MONITORING FIRMS
8
AGENCY THEORY AND MONITORING
selling securities, while the true nature of the return distribution of the
project is unknown to the outside market. Management possesses
valuable information about the project that is unavailable to the market.
If this information were revealed fully to the market, the market would
value the project at V . Otherwise, the market is unable to distinguish
a
this project from another less profitable project with a value V . This
b
asymmetry may be corrected, at a cost, through various ‘signalling
mechanisms’. In the absence of an unambiguous signal, however,
management will obtain less for securities sold than their ‘fair value’
reflected in the true nature of the project A. The difference between the
‘fair’ price and the actual price is the agency cost associated with
informational asymmetry. It exists for the issuing of debt as well as
new equity securities, provided that there is a differential probability of
bankruptcy for the two projects. In their seminal paper, Jensen and
Meckling (1976) proposed agency costs as a key tool in evaluating
designs of a principal-agent relation. Agency costs were defined as the
sum of:
9
THE ROLE OF BANKS IN MONITORING FIRMS
The second case concerns the largest part of the agency literature:
adverse selection models. Adverse selection can be represented by
Figure 1.2.
Kose and Kalay (1982) illustrate adverse selection by analysing the
conflicts between two groups (stockholders and bondholders) concerning the
riskiness of company projects. They derive an optimal set of contractual
arrangements which minimise the cost of this conflict. In that sense, these
authors come back to the idea of Myers (1977), who has argued that
stockholders who control the firm could attempt to transfer wealth from the
bondholders, even by rejecting profitable investment projects. In other words
they can choose policies which reduce the market value of the firm. Myers
has indicated that self-imposed constraints on dividend pay-outs by
stockholders are a possible solution to the problem of under-investment.
The debt contract between a bank and the manager of a firm defines the
adverse selection problem. According to the recent literature, such as
Laffont and Freixas (1988), Gale and Hellwig (1985) and Diamond (1984),
the assumption of asymmetric information leads the bank to prefer credit
rationing, to an equilibrium between supply and demand at a specific
interest rate. The agency relations between a bank and a firm prevent the
lender from distinguishing sound creditor risks from bad ones. The
requirement of a guarantee can solve such problems and enables the lender
to separate good borrowers from the bad ones. In this context Deshons and
Freixas (1987) establish two types of loan system:
• the optimal separating contracts, where good credit risks are identified;
• the rationing credit solution, where neither guarantee nor interest rate
make the selection role possible since the bank is unable to take the
surplus of its loan customers.
10
AGENCY THEORY AND MONITORING
11
THE ROLE OF BANKS IN MONITORING FIRMS
shareholders, who seek dividends, the managers, who must signal the value
of their organisation by diversifying their portfolio, and the banks, which
grant credit and want to avoid losses if the firm goes bankrupt.
To conclude these first two points, it seems that the application of agency
and signalling theory to finance decisions is concerned with the influence of
conflicts between shareholders and managers over the value of the firm, and
the resolution of these conflicts by the use of debt. The latter involves an
incentive scheme for performance. The debt increases the bankruptcy risk,
which is a threat sufficient to encourage the manager to administer
efficiently. Bankruptcy will mean for the manager not only loss of his work,
but also being held responsible for the situation.
In these models, debt is defined by an optimal contract for the firm,
and the optimal structure of the enterprise is a combination of debt and
stocks. So, the financial parameters justifying the limits of debt policy
can be deduced by analysis of the shareholders’ and managers’ points
of view (e.g. level of risk of the project, expected profit, etc.).
However, these theories have a certain number of limitations. In
particular, the same financial indicators have specific roles in each
approach. The debt level constitutes a very manipulable variable for the
administration of the firm. It signals the quality of an enterprise for
Ross (1977). It is the origin of the agency cost for Galai-Masulis
(1976) and Jensen and Meckling (1976). The results of these analyses
do not seem to be very coherent: an increase in debt or in the part
attributed to the manager and shareholders in the capital induces an
increase in the market value of the firm, according to the signalling
models. But the difficulties due to agency relations then become more
and more pronounced. Myers (1977) has taken an interest in this
problem. He explains that three financial parameters must be taken into
consideration at the same time: the dividend distribution, the amount of
investment and the level of debt.
Agency costs appear there as one of the essential elements of a new
kind of signalling model. For Myers, the existence of debt for an
enterprise can generate a sub-optimal level of investment: the nature of
equity and the financial structure can be modified such that the investors
do not maximise the value of the firm but the value of the stocks, to the
prejudice of debt.
As indicated above, agency theory aims to explain the conflicts of interest and
differing information that could exist between the different partners of the firm:
managers, shareholders and lenders. Every partner determines the optimal level
of debt according to his own interest. The manager wants to find the amount of
12
AGENCY THEORY AND MONITORING
funds he thinks sufficient. For the shareholders and the bank, investment is a
way to realise profit. However, the latter actors bear different costs. The
shareholders give the manager financial compensation for this work, while the
banker pays interest costs for granting the credit (Laffont and Freixas 1988).
In this section, possible areas of common interest between these three
actors in the financing of the firm will be discussed. In particular the
following question will be asked: are these different objectives sufficiently
compatible such that the common determination of debt for the firm might
be possible? Are there situations where the three partners are led to define
jointly the very price to pay, so that the firm could finance its productive
activities through an optimal level of debt?
When several economic agents must take a decision in which they are all
concerned (for example, the conclusion of a debt contract between bank and
manager, preserving the solvency of the enterprise and the wealth of the
shareholders) the economic argument is both normative (what is the optimal
price of debt for a firm?) and strategic (which price results from the
confrontation of these different interests?). In fact the shareholder is likely
to be in favour of debt which gives him more profit and decreases his level
of risk-sharing. However, if the manager invests in high-risk projects, the
risk of bankruptcy increases. Lars Stole (1990) has considered the
contracting problem facing multiple principals, each of whom wants to
contract with the same agent. This has been termed the common agency.
Common agency can either be delegated or intrinsic. Under delegated
common agency, the choice of contractual relationship is delegated to the
agent, who can choose whether to contract with both, one or none of the
principals. Alternatively, when common agency is intrinsic, the agent’s
choice is more limited: the agent can choose only between contracting with
both or with neither.
Consider a contracting environment with two principals and one agent. A
first possibility is that each principal wants to share the profit of a given
outcome which the agent produces. Each agent is represented by a type
which describes the characteristic of the projects undertaken by the manager.
A direct revelation mechanism incites the agent to announces his type to
each principal separately. If the agent contracts only with one principal we
return to the literature defining contracts between either shareholders and
manager or between bank and manager. If the agent contracts with no one,
as he has not enough money to realise the project, he cannot carry it out.
The only interesting case is where all parties contract:
• Stage 1 The principals announce the contract which they wish to sign
with the manager (agent). A contract specifies a project, a reward
scheme for each configuration of the market (principal-agent) and the
credit to be granted. We assume that offers between principals and agent
are private information.
13
THE ROLE OF BANKS IN MONITORING FIRMS
• Stage 2 The manager of the firm (agent) announces which of the offered
contracts he is willing to accept. The announcement is public
information (each agent must inform the principals whether he wishes to
deal with that project exclusively or in conjunction with projects of
other firms). Public observability of the announcement is not essential,
but simplifies the description of an equilibrium.
• Stage 3 Each principal selects the manager which is willing to
undertake the project and to accept the proposed contract. At the same
time, principals announce the amount of capital they grant for its
realisation. These choices are observable.
Assumptions
First, all players are risk-neutral. While risk-neutrality on the part of the
principal is not essential, a risk-neutral agent is indispensable in what
follows. For example, as mentioned in stage 2, a manager can undertake
several projects with differing levels of profitability, which enables him
to spread risk. Second, principals are allowed to condition the terms of
contracts upon the market. This feature of our model illustrates an
important aspect of agency delegation. Since a principal’s decision to
abandon a common agent (manager) constitutes an observable change in
market environment, competitors may renegotiate their contracts in
response. 3
Stage 1 of our model provides a simple way to represent this process of
reactive renegotiation: the principal simply announces an amount of funds
and an incentive scheme for every possible market configuration. Note,
however, that, as long as a principal retains the same agent, other principals
cannot respond directly to changes in its contracts since contract offers are
private information. Consider now the model described above, but modify
the description of the agency delegation process as follows. In stage 1, each
principal makes one contract offer (credit incentive scheme) to the same
agent (the manager of the firm). In stage 2 this single agent either accepts
both offers or neither. If he rejects, principals earn zero profit and the game
ends. In this simple model, common agency is intrinsic in the sense that the
agent has no alternative to the pre-specified delegation. Bernheim and
Whinston claim that:
14
AGENCY THEORY AND MONITORING
1 why debt was relied upon as a source of capital before debt financing
offered any tax advantage relative to equity;
2 why shares are issued when financing a project.6
15
THE ROLE OF BANKS IN MONITORING FIRMS
Some authors who have considered this question have observed that
financial systems differ significantly between countries and across periods,
and the question arises how these differences between financial systems
have affected the functioning of the different economies. In this section I am
going to concentrate on two case studies, Japan and Germany, in order to
contrast the results obtained for these two countries with those relating to
the Crédit Mobilier in Chapter 2. As a first illustration of this concept, let us
consider the paper of Mayer (1988). The author suggests there that there are
systematic differences in performance between financial systems in which
banks play a prominent role, and those financial systems in which banks are
less prominent. He explains these differences in performance in terms of
differences in the mechanisms that reduce or eliminate moral hazard
between entrepreneurs and financiers. Specifically, he suggests that the more
bank-oriented systems of Germany and Japan involve greater commitment
on the part of the firm and the bank to a long-term relationship, which
allows them to enjoy the benefits of long-term contracting.
In this section, the following issue is considered: in what sense is ability
to support long-term relationships the clue to assessing financial
institutions? The focus of the discussion will be on financial institutions
(more specifically, the role of banks in the provision of finance to firms),
rather than financial instruments.
Theoretical framework
The manager-debtholder literature has two sub-literatures. First there is the
issue of the manager-bondholder conflict. There are numerous theoretical
discussions of the role of bond convenants in controlling the manager’s and
(or) shareholders’ incentive to increase the risk of the firms. This literature
has treated debt mostly as a homogeneous instrument creating incentives, in
direct conflict with those associated with equity financing. Second, there is
the borrower-lender literature which has developed from the simple
examination of the incentive of the borrower to default on a bank loan, to
the role of the bank as the delegated monitor of a portfolio of borrowers,
i.e., a monitoring role for the bank’s depositors.7 These latter models have
served to strengthen the argument that there is a unique information and
monitoring role of bank loans and that the phenomenon of credit rationing is
due to the nature of information asymmetry in the market.
The argument discussed here is richer than the traditional literature
briefly presented above. Essentially based on the model of Devinney and
Milde (1990), it incorporates the role of ‘inside debt’. Inside debt, as
16
AGENCY THEORY AND MONITORING
17
THE ROLE OF BANKS IN MONITORING FIRMS
• in the late nineteenth century, a hundred years after the beginning of the
industrial revolution, the adoption of up-to-date technology required
large capital outlays; and
• the failure of some countries to participate in the earlier stages of
industrialisation meant there was a lack of accumulated funds that could
be used to finance these outlays.
18
AGENCY THEORY AND MONITORING
The discussion of the advantage of bank finance over market finance implies
that financial systems are institutions which reduce or eliminate problems of
moral hazard or asymmetric information between firms and financiers.
Financiers typically have less information about firms than entrepreneurs or
managers. Moreover, investors are subject to various types of moral hazard:
moral hazard concerning managerial effort, moral hazard concerning the
19
THE ROLE OF BANKS IN MONITORING FIRMS
On the basis of these two propositions, Diamond shows that under certain
circumstances incentive-efficient allocations cannot be made without
intermediation.
Diamond’s notion of financial intermediation as delegated monitoring (or
delegated control) is closely related to Gerchenkron’s account of bank
involvement in firms at the early stage of industrial development, although
this is not explicit in Gerschenkron’s work. As emphasised by Mayer, bank
initiative and bank participation in entrepreneurial planning may be a way to
obtain enough information and control to reduce the moral and
informational hazards of finance to a tolerable level. We may therefore look
at the imperfect information approach to financial intermediation as the
theoretical basis for Gerchenkron’s view that banks were needed to provide
outside finance during the early stages of industrialisation in Germany, when
capital was ‘scarce and diffuse’ and ‘the distrust of industrial
activities…considerable’. In Diamond’s terminology, Gerschenkron’s
assumption must have been that the sum of the monitoring costs of the
banks and direct agency costs of bank deposits was less than the agency
costs of direct finance, perhaps even that the agency costs of direct finance
were so high that this was not a genuine alternative at all.
According to this approach, supervision seems to be profitable for both
the bank and the shareholders, which stand to benefit from higher profits by
reducing their agency costs. It will be interesting to analyse which type of
contract is induced to exert this control. In other words, are long-term
commitments necessary to control firms?
20
AGENCY THEORY AND MONITORING
21
THE ROLE OF BANKS IN MONITORING FIRMS
22
AGENCY THEORY AND MONITORING
23
THE ROLE OF BANKS IN MONITORING FIRMS
This first analysis describes the relationships that exist between the manager
of a firm, its shareholders and the bank which finances the investment
24
AGENCY THEORY AND MONITORING
25
2
2.1 INTRODUCTION
The aim of this work is to gain, through empirical research, an insight into
the role of banks in the financing of enterprises during the period 1850–
1913 in France. Some authors (e.g. de Long (1989) and Ramirez (1992))
have already applied this analysis in the United States.
The credit which a bank gives to a company enables it to exert an
influence over the company. 2 J.Bradford de Long has illustrated the
role of investment bankers in his article ‘Did J.P.Morgan’s men add
value? A historical perspective on financial capitalism’. In the period of
which he writes, financiers possessed strong voices in corporate
management. The author illustrates his point of view by focusing on
the J.P.Morgan Company. J.P.Morgan and his partners saw themselves,
as did other participants in the pre-WWI securities industry, as
fulfilling a crucial monitoring and signalling intermediary role between
firms and investors, in a world where information about the underlying
values of firms and the quality of management was scarce. De Long
stresses that investment bankers played an important role, in that they
took great care to make sure that the firms they watched had the right
managers. This analysis fits our hypothesis that the bank can control
the management of a firm and motivate it to reveal truthfully the
information about the project it proposes to undertake.
Another line of research has been made in the German context. The
roles of four banks—the Deutsche, the Diskonte Gesellschaft, the
Dresdner, and the Darmstädter—have been studied, producing the same
26
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
conclusion. Bradford de Long and Marco Becht (1992) have used price
and dividends series to test the volatility of the German stock market.
In particular, they examined whether the role played by the German
Großbanken in the pre-WWI stock market might have been the cause of
the low comparative volatility of the German stock indices before 1914.
These ‘Great Banks’ were founded in deliberate imitation of the French
Crédit Mobilier. There has been much discussion about the origin of
the early German credit banks, and especially to what extent the Crédit
Mobilier served as a model. The first significant joint-stock bank was
created in 1848: the Schaaffhausen’schen Bankverein. During the crisis
of the 1860s’ recession it faced a liquidity problem. Because of the
bank’s importance and the strained atmosphere of the times, normal
liquidation procedures did not occur. The Darmstädter was a direct
result of the French Crédit Mobilier. It counted among its founders
Abraham Oppenheim, who had been co-founder of the Crédit Mobilier.
Much of the initial capital of the bank was supplied by the Péreire
brothers.
These banks had access to considerable capital with which to
provide finance to corporate enterprises. They contributed to the
intensity of the boom of the 1850s. Despite the crash in 1857 and the
depression of the 1860s, most of the banks survived. The early German
banks remained permanently tied in with the companies which they
promoted, through representation on boards of directors and direct
participation in ownership. How this cautious approach can be
reconciled, at least in theory, with the role the banks wished to play in
the development of industry, is clear from the Schaaffhausen’schen
Bankverein for 1852
The case of the Crédit Mobilier will be analysed to see if, as in the
American and German contexts, bankers possessed strong voices in
corporate management. From 1853 to 1913 this bank took part in the
financing of firms. In fact, two ‘Crédit Mobiliers’ existed. The first, founded
by the Péreires, went bankrupt in 1867. After its bankruptcy it was
reorganised, but it went bankrupt again in 1900. It reappeared in 1903 as the
27
THE ROLE OF BANKS IN MONITORING FIRMS
28
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
29
THE ROLE OF BANKS IN MONITORING FIRMS
30
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
Statistics on the issue of shares and the interest rate on the financial
market give an idea of the fluctuations on an aggregate level. 7 These
statistics serve as a reference point for the analysis of the relations between
the Crédit Mobilier and the companies with which it was in contact.
Nevertheless, operating only with its own capital and deposits, the Crédit
Mobilier performed prodigious feats of industrial promotion and financial
manipulation. In its first major operation, it subscribed to 40 million francs
in the Crédit Foncier’s bonds, with which it maintained close ties in the
early years.
In its first full year of operation the Crédit Mobilier financed, by
making advances and underwriting bond issues, the Midi railway as
well as the newly chartered Grand Central and the French Eastern,
Mulhousen and Strasburg-Basle railways. The Crédit Mobilier also
secured the merger of three short railways in the Loire coal basin with
the Grand Central, reorganised the Loire coal industry and promoted
the Darmstädter Bank. It played a leading role in the mergers which
took place in 1855 in the French Western (railway), and promoted or
financed the railways of the Pyrenees, Dauphine, the Ardennes and
several shorter lines. It created, with generous government subsidies,
the Compagnie Générale Transatlantique (French Line), the Compagnie
Générale des Omnibus (Paris), and the Société de l’Hôtel et des
Immeubles de la Rue de Rivoli (subsequently the Compagnie
Immobilière).
The Crédit Mobilier founded by the Péreires played a great role in French
industrialisation. In every company financed by this bank, a member of the
Crédit Mobilier’s board of directors was represented in the shareholders’
assembly. The annual report of April 1860 is very explicit on this subject:
31
THE ROLE OF BANKS IN MONITORING FIRMS
32
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
33
THE ROLE OF BANKS IN MONITORING FIRMS
Table 2.3 Fluctuations of share prices for affiliated companies between 1855 and
1856
The fall of the Crédit Mobilier was essentially caused by the attempt to
salvage the Société Immobilière. This company’s eagerness to participate
fully in the extension of Paris led its managers to undertake huge
investments with poor potential.
From 1864–5 the Crédit Mobilier had to confront serious difficulties
arising from the Banque de France’s decision to increase interest rates. The
Crédit Mobilier decided to double its capital, to support advances for the
Compagnie Immobilère. But in June 1866 its situation became worst and the
fall of the share prices seemed to be the consequences of the depreciation of
its portofolio. The Péreires redoubled their efforts to obtain cash. It was said
at the time that a loan would be granted to the Crédit Mobilier by the city of
Paris, but the deal was never realised. By the end of 1866, the Péreires were
contemplating the possibility of a merger between the Mobilier, the
Mobiliario Espanol and the Immobilière, with a loan from the Caisse des
Consignations (an agency of the government), but they did not succeed. In
addition to these difficulties, the Péreires faced a stockholders’ dispute
34
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
concerning the increase of the bank’s capital. The Crédit Mobilier’s true
situation could no longer be concealed. In March 1867, for the first time in
its existence, the prices of its shares fell below par.
The Péreires were forced to resign from both the Crédit Mobilier and the
Immobilière. Between 1867 and 1870, little information was available. The
Crédit Mobilier went bankrupt, Emperor Napoleon III, who had supported
the Péreires, fell and France had to face a war with Germany. In 1871,
Baron Haussmann made a deal with the bank for the repayment of its debt.
This involved formally dissolving the old company, and immediately
forming a new one. The new bank, called the Société de Crédit Mobilier
Français, was to take over the assets of the old Crédit Mobilier.12 The impact
of this new bank on the financing of French industry was so small that it
disappeared from the economic literature. The bankruptcy and loss of
resources meant it played only a secondary role.
After the fall of the Péreires in 1871, Baron Haussmann, an old member of
the board of directors, became President of the new Crédit Mobilier.
However, as mentioned previously, the bank had lost its earlier pre-eminent
role in the financing of French industry. The first aim of the new bank was
to solve the difficulties of the Compagnie Immobilière. The capital of the
company was divided in two parts, between Paris and Marseille. Part of this
credit came from the Crédit Foncier, which had contributed to the financing
of the Compagnie Immobilière: 55 million francs came from the credit of
the Paris division and 32 million from Marseille. Floating assets were
liquidated. The Crédit Mobilier reduced its holdings of the Compagnie
Immobilière’s debt to 51 million francs by selling part of its bonds and
issued new assets for an amount of 76 million francs. Half of these assets
were bonds with fixed revenue (5 per cent) and half were bonds with
variable revenue to the amount of 5 per cent of the capital. The liquidity
problems of the Compagnie Immobilière were solved by the partial
liquidation of its debt. The Crédit Mobilier became a firm with a par value
of only 40 millions francs, undertaking more projects abroad than in France.
In 1875, Emile d’Erlanger, another member of the old board of directors
(of the first Crédit Mobilier) became President. During this period the bank
had relations with the Bank of Brussels and the Spanish government. No
balance sheet was published in the reports to the shareholders’ assembly,
despite the fact that they had been specifically requested by the
shareholders.
In August 1876, Baron Emile d’Erlanger submitted his resignation and
Charles Wallut, the bank’s Vice-President, became President. The Crédit
35
THE ROLE OF BANKS IN MONITORING FIRMS
Mobilier continued to start new ventures with foreign companies, but little
information was published relating to its activities within France. The bank
participated in the Italian Ice House Society, as well as contributing to an
Algerian government loan.
In June 1881, Eugène Péreire, founder of the first Crédit Mobilier, was
nominated Administrator of the bank. He proposed to add a bank service to
the Compagnie Transatlantique. The Banque Transatlantique was founded
with a capital of 50–60 million francs, of which one quarter was to be given
immediately to the Company Transatlantique. Twenty thousand Comptoirs
Maritimes shares were absorbed by the new bank. The Crédit Mobilier took
10,000 shares in this new society. In August 1881, M.Péreire was appointed
to represent the Crédit Mobilier on the syndicate which was formed to issue
the Banque Transatlantique shares. The Crédit Mobilier, with the Romanian
government, undertook the constitution of a Romanian Crédit Mobilier.
Economic relations were developed with the United States and Spain for
railway construction.
On 4 November 1882, M.Péreire raised the issue of the participation of
the Crédit Mobilier in the profits of the Compagnie Transatlantique. He
thought that the idea of a combination of capitalisation or repurchase of the
right to the profit sharing should be examined. On 3 February 1883, Péreire
submitted his resignation. This was followed, on 29 June that year, by the
Crédit Mobilier asserting its rights to the benefits of the Compagnie
Transatlantique. It was believed that the Crédit Mobilier should have 25 per
cent of these profits and the shareholders 8 per cent. Subsequent court
proceedings resulted in the Compagnie Immobilière being ordered to pay
367,000 francs to the Crédit Mobilier for 1881 and 1882.
In 1886, the Crédit Mobilier was a company with 30 million francs of
Sources: Crédit Mobilier: Procès Verbaux des Assemblées Générales; Procès Verbaux
du Conseil d’Administration
36
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
capital. Its financial situation got worse and worse, until it became
completely bankrupt in 1902. One consequence was that, in 1898, funds
were cut to two main borrowers, causing the Puerto Rican railway to stop
and the Rhone railway line to be liquidated. In 1901 the Crédit Mobilier’s
liquidation account was as shown in Table 2.4.
The Crédit Mobilier was anxious to prove that its portfolio was
performing and it took part in business which was small and easy to realise.
Between 1903 and 1904 the Crédit Mobilier began to develop: its capital
reached 10 million francs and, the year after, 25 million francs. During this
period the bank had business relations with the large financial house of
Thalmann and Co. and had just concluded a South American business deal.
The bank was then capable of entering, either alone or with the
participation of other companies, more suitable financial operations (e.g. the
Serbian national debt). The Crédit Mobilier maintained relationships with
provincial banks to guarantee full realisation of share transactions. The
report of 1911 states:
37
THE ROLE OF BANKS IN MONITORING FIRMS
38
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
They argued that the central bank should intervene for illiquid but solvent
banks. Otherwise supporting failed companies would damage healthy
competitors and weaken the discipline of a market economy.15
For Thornton, the Central bank’s responsibility in times of panic was
to act as lender of last resort, providing liquidity to the market and
discounting freely the paper of all solvent banks, but denying aid to
insolvent banks no matter how large or important (Humphrey 1975,
1989). In other words, the lender of last resort must prevent illiquid but
solvent banks from failing. 16
At the other extreme, Charles Goodhart (1985, 1987) advocates
temporary central bank assistance to insolvent banks. He argues that the
distinction between illiquidity and insolvency is a myth, since banks
requiring lender of last resort support because of ‘illiquidity will in most
cases already be under suspicion about…solvency.’ Furthermore ‘because
of the difficulty of valuing [the distressed bank’s] assets, a Central Bank
will usually have to take a decision on last resort support to meet an
immediate liquidity problem when it knows that there is a doubt about
solvency, but does not know just how bad the latter position actually is’
(1985:35).
Solow (1982) is also sympathetic to assisting insolvent banks. He argues
that any bank failure, especially a large one, reduces confidence in the
whole system. To prevent a loss of confidence caused by a major bank
failure from spreading to the rest of the banking system, the central bank
should provide assistance to insolvent banks. However, such a policy
creates a moral hazard, as banks respond with greater risk-taking and the
public loses its incentive to monitor them. In the event, the Banque de
France did not rescue the Crédit Mobilier. Its reasoning followed
Thornton’s argument, according to which:
39
THE ROLE OF BANKS IN MONITORING FIRMS
Hence the non-intervention of the Banque de France was profitable for the
whole system, because an insolvent bank disappeared.
From an historical and human point of view, a simple study of the main
shareholders of the Banque de France sheds further light on this non-
intervention. During the period, the Rothschild family, who disliked the
Péreires for both personal and professional reasons (and may have been
jealous of the power the Crédit Mobilier had obtained in the financial
market) had contributed to the decision of the ‘Banque de France’ not to
salvage the Crédit Mobilier. The only intervention that had occurred was to
prevent a too sudden and huge fall of share prices on the stock market,
which would have provoked a financial crisis for the whole economy. As
regards this argument, it will be interesting to analyse the impact of the
Crédit Mobilier’s bankruptcy on the share prices for both affiliated and non-
affiliated companies.
This section presents and discusses all the data sets that will be used in the
empirical tests explained below. Three points are developed: the first aims to
elicit the general context of the French economy and to analyse in more
detail the situation of the financial markets; this will constitute the basis of
our excess volatility test.
The second and third parts describe the data available for companies that
were affiliated with the Crédit Mobilier and independent ones; this will
support our statistical research relative to the examination of the impact of the
Crédit Mobilier’s bankruptcy for both sub-samples and for the whole sample.
As Bradford de Long and Marco Becht (1992) did for the United States
and Germany, I intend to analyse the excess volatility of the French
stock market from 1852 until 1914. In this respect some general ideas
about the French economy from 1853 to 1914 are given. 17 More
specifically, general figures on share prices, market prices, interest rate
and dividend are provided.
The aggregate data are relatively simple. The money stock was stable.
The huge saving rate compared to a reasonable demand maintained
interest rates at a moderate level over the period: between 3 per cent and
40
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
4.5 per cent (Loutchitch 1930). The rate of discounting assets with fixed
revenue varied between a minimum of 3.22 per cent in 1897 and 1898
and a maximum of 3.73 per cent in 1913. The official discount rate
decreased to 2 per cent in 1896 and 1897. Its value was around 3 per
cent from 1900 to 1910. This low return on credit gave the banks an
incentive to invest abroad. Some explanations are required to explain
how the indices are calculated. To construct the industrial, transport and
trade series, fifty four separate assets have been taken into consideration
for the years 1895 to 1913 and twenty four assets for the years 1870 to
1895. Amongst these were some issued by companies involved with the
Crédit Mobilier. From 1895 to 1913 these were:
• Electrometallurgie de Dives;
• Chemin de Fer PLM, Nord, Orleans, Midi;
• Compagnie Générale Transatlantique;
• Société Parisienne d’Eclairage.
From this series, the impact of the Crédit Mobilier’s bankruptcy on the
level of production can be analysed. If the Crédit Mobilier had been a
‘Great Bank’, the impact of its bankruptcy on affiliated companies could
have induced a relative effect for the sectors to which these firms belonged.
As, for a large number of sectors (e.g. the railways), the Crédit Mobilier was
an active partner in the financing of companies, it will be interesting to
present some figures to support the argument.
Table 2.5 compares market product prices by sector on an aggregate
level. The sectors I have chosen are those in which the Crédit Mobilier
Source: Lenoir (1919:87–8) basis average of the market prices for 1901–10
41
THE ROLE OF BANKS IN MONITORING FIRMS
• Did the Crédit Mobilier stabilise the ratio of market value to dividends
of affiliated firms?
• Did the Crédit Mobilier increase the market value compared to book
value of affiliated firms?
In addition, I want to study the power of the Crédit Mobilier among the
banking sector, its policy towards investment and the importance of an
affiliation with the bank as regards credit policy.
P.Dupont Ferrier in his book Le Marché Financier sous le Second Empire
provides a list of the societies founded by the Péreires between 1853 and
1867. These enterprises were:
42
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
The bonds issued by the firm were guaranteed by its capital. The
received capital during the year could not be greater than the capital
realised (Cameron 1991:142). Plenge (1903) gives the data on the
Crédit Mobilier’s shares throughout its existence, their value in 1867
(the year when it went bankrupt) and the distribution of dividends (see
Appendix A).
The new Crédit Mobilier (1903–14) was initially involved with the
Société métallurgique de Montbard-Aulnaye, in which the bank was
largely represented, and with the Société d’Electro-métallurgie de Dives.
The Crédit Mobilier also founded the Société Française des Machines-
Outils. In varying degrees, the Crédit Mobilier participated in the
Compagnie Parisienne de Distribution d’Électricité and in the
reorganisation of the Compagnie Générale des Omnibus de Paris. It
contributed to the development of the Société de Raffineries Say and to
the expansion of the Société de l’Éclairage Électrique. The bank also
concluded contracts abroad. In 1909–10 it participated in an Argentinian
government loan whose interest rate was 4.5 per cent and in the Santa Fe
business affair (interest rate 6 per cent).
In summary, the firms in which the Péreires were interested were the
railway companies, l’Union, the Compagnie Transatlantique and the
Compagnie Immobilière de Paris. Two societies were involved with both the
Péreires and with the Crédit Mobilier Français: the Société des Omnibus de
Paris and the Compagnie d’Eclairage et de Chauffage par le Gaz. Firms
involved with the new Crédit Mobilier included the Société des Machines
Outils, the Société Montbard-Aulnoye, the Electro-metallurgie de Dives and
the Raffineries Say.
In contrast to the first Crédit Mobilier founded by the Péreires, the
second one was not always represented on the boards of firms with which
it was involved. In the Raffineries Say, however, the Crédit Mobilier was a
financial partner of the company, while with the Société des Machines
Outils, the Société Montbard-Aulnoye and the Electro-Metallurgie de
Dives the Crédit Mobilier was strongly represented on the boards of
43
THE ROLE OF BANKS IN MONITORING FIRMS
directors, and sometimes also the principal shareholder, as with the Société
des Machines Outils.
However, the new Crédit Mobilier’s functioning was not very different
from that of the Péreires, particularly when it was well represented on
boards of directors. The close relations that the Crédit Mobilier
maintained with the societies with which it was involved enabled it to be
not only a financing establishment but also an active partner in the
realisation of projects.
As advances were granted to the societies where the Crédit Mobilier was
officially represented, the risk of bankruptcy was increased. In this case we
can ask the following questions:
• Did association with the Crédit Mobilier increase the value of shares
compared to book value?
• Did association with the Crédit Mobilier reduce the variance of the
price dividend ratios?
• Did association with the Crédit Mobilier always facilitate loans and
advances to the companies? If so, in what measure?
This data set supports the hypothesis that companies not affiliated to a bank
tend to be liquidity constrained (and face a larger gap between external and
internal costs of funds) than those which are. Bertrand Gille (1968) provides
us with a list of companies which were affiliated with the Crédit Lyonnais
for the financing of their projects. These companies were
• Alais;
• Schneider;
• Marine;
• L’home;
• Terrenoire;
• Denain;
• Fourchambault;
• Maubeuge;
• Givors;
• Firminy;
• Franche Comté;
• Chatillon.
44
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
(2.1)
45
THE ROLE OF BANKS IN MONITORING FIRMS
price level, d represents the dividends divided by the price level and r is
chosen to represent either the an arbitrary real required rate of return or a
risk premium paid on assets plus the risk-free rate minus the inflation rate.
Here r is similar to the value taken by Becht and de Long (8 per cent) to
facilitate the interpretation of the result. Other values for r between 5 per
cent and 10 per cent would not have changed the shape of the graph in
Figure 2.1.
In the French case I have calculated the fundamental price using
Dessirier’s series for the share prices and the dividend and Lenoir’s series
for the general level of prices. All the series constructed are deflated by
the GNP deflator from Lenoir. The same methodology used by Bradford
de Long and Becht (1992) will be followed. As the general price series,
the share price series and the dividends series do not refer to the basic
year in the construction of the indices, normalisation of all these prices is
necessary to avoid any bias problems in the test. Moreover, the dividend
series presented in Dessirier’s paper was not used as such. We evaluated
the ratio as dividend.100/share price. This series corresponds to a constant
term times the ratio of dividend index to price index. After having
normalised the dividend, a calculation of the volatility ratio, that is the
price dividend ratio, is made in order to construct the test. Figure 2.1
provides individual surveys of the behaviour of prices, dividends, and
perfect foresight fundamentals which are obtained by computing equation
46
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
(2.1). They plot for the period the logs levels of prices and the log ex post
perfect-foresight fundamental.
Shiller’s test says that:
47
THE ROLE OF BANKS IN MONITORING FIRMS
Certain bankers of the first half of the nineteenth century (and more later
also such as Laffite, or, more importantly, the Péreires) thought and said that
the progress in the banking system, in feeding payment and credit, enabled
growth and avoided ‘commercial crises’. The facts contradicted their
optimism. The promotion of banking and the modern form of credit did not
prevent cycles. They furnished, on the contrary, a source of new turmoil: the
rise of credit expansion and the confirmation of depression due to the
increase of banking operations.
48
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
• The first one concerns the effective power of each bank. In particular
the Crédit Mobilier and the Großbanken did not have similar powers
with regard to screening companies’ projects. The German Great Banks
were more cautious and evaluated the level of risk for any potential
investment project.
• The second deals with the ability of the Crédit Mobilier to supervise the
managers of the firms which whom it was involved. The differences
described in the text between 1870–95 on the one side and 1895–1914
on the other side give a hint of the different policy which had been
exerted by the German banks and the French Crédit Mobilier.
• The third refers to the economic situation in France during the
period and, more specifically, the difficulties faced by the banking
system because of its innovative character as evidenced by Saint
Simon’s theory.
Table 2.6 Volatility tests for the French, German and American stock markets
between 1870 and 1914
49
THE ROLE OF BANKS IN MONITORING FIRMS
extremely flat before 1895 (some values are identical for subsequent
years) and reported yearly after 1895, the ratios are evaluated for the
whole sample and for 1895–1914.
The first line of Table 2.6 shows the volatility of log prices (p) around
the log of the perfect foresight fundamental (p*–p) calculated using an 8
per cent per year real discount. The second line shows the volatility of
the log price-dividend ratio (p–d). The third line shows the volatility of
the log ex post perfect foresight fundamental to dividend ratio (p*–d). If
prices are more volatile relative to ‘naive’ forecasts than perfect
foresight fundamentals are, relative to the naive forecast, investors could
and should have constructed a better forecast: a weighted average of the
market price and the naive forecast would have generated smaller
forecast errors. Thus the values in the second line of the table, the
volatility of the price-dividend ratio, should under the efficient market
hypothesis be smaller than the values in the third line, the volatility of ex
post perfect foresight fundamental. For France it is true after 1895 as for
Germany. Between 1870 and 1914 the situation is more similar to the
United States. Line four reports this volatility ratio. Line five of the table
calculates another volatility ratio. The volatility of the perfect foresight
fundamental about the actual price should be less than the volatility of
the perfect foresight fundamental about the naive forecast. For both
periods the price and the dividend are almost exactly equally good
forecasts of the perfect foresight fundamentals.
A final implication of the efficient markets hypothesis is that the two
ratios of lines four and five—the sum reported in line six—should add
up to one. If not (as is the case of France for the whole sample, and the
United States) the log difference between price and the perfect foresight
fundamental (p–p*) is correlated with the log price-dividend ratio (p–d).
Profits could have been earned by trading on this correlation of the
price-dividend ratio and value relative to price. Surprisingly, between
1895 and 1914 the situation of France is similar to that of Germany. The
analysis of the results obtained for France leads us to the following
conclusions:
1 Prices are volatile relative to dividend over the period 1870–1914. Prices
are much more volatile than perfect foresight fundamentals relative to
the naive forecast. Thus, tests based on market volatility ratios shows
traces of excess volatility in the French stock market before the First
World War. However, between 1895 and 1914 volatility tests present
strong similarities between France and Germany
2 The situation described by the French stock market is much more similar
to that of the United States for the whole sample than that of Germany.
The symbol ‘ used in Dessirier’s data should lead to a bias which gives
opposite results for the whole sample and the period 1895–1914.
50
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
51
THE ROLE OF BANKS IN MONITORING FIRMS
sought also to support the prices of the securities they issued. For Riesser,
this support was not an attempt to produce an artificial market or to swindle
the public. These authors wonder if the Großbanken had sufficient resources
to keep asset prices close to their own internal estimates of fundamentals. As
the Großbanken were by far the largest actors in the German economy, an
affirmative answer can be given to this question. In the United States the
situation was similar to that in Germany: the banks occupied a powerful
position. To illustrate this argument, let us just quote Karl Erich Born
(1983:93):
In France such a situation did not exist except for the ‘Haute Banque’
and more specifically the Rothschilds, whose aims was to fight the Péreires.
The Rothschilds found the Péreires’ investment policy too risky to consider
them as real bankers. On the other hand, they saw them as competitors they
had to force out of the market. Evidence of this lack of supervision will be
provided in the following section by analysing the impact of the Crédit
Mobilier’s bankruptcy on the financial market.
(c) Conclusion
52
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
1 the effect of price level and dividend on share prices for the Crédit
Mobilier’s firms;
2 the correlation between the failure of the bank and the fluctuation of
share prices for the affiliated companies;
3 an evaluation of the association with the Crédit Mobilier between 1866
and 1868.
53
THE ROLE OF BANKS IN MONITORING FIRMS
Table 2.7 Average asset prices for affiliated firms
Notes
a
When the share price increased this is indicated by a minus sign (i.e.
negative loss).
*
This must be considered as the loss per share before the reorganisation of
the Crédit Mobilier and the issue of the new shares. After the fall of the
Crédit Mobilier the Cie Immobilière went bankrupt: so the same comments
as those made for the Crédit Mobilier apply in that context.
**
After the bankruptcy of the Crédit Mobilier the Compagnie Transatlantique
chose to reduce its partnership with this bank. For example, in 1864, the
amount due to the Crédit Mobilier was 10,126,461.08 francs. The company
was supposed to give 25 francs of dividends deductable from a reserve fund
in case of insufficient profit over the year. In 1867 the dividend was only 15
francs per share and the Crédit Mobilier became a debtor of the company
for an amount of 4,514,610.78 francs. In 1868, M.Dollfuss became President
of the Compagnie Transatlantique; M.Péreire had to resign.
The aim of this section is to see if there was any marked effect on price
level and dividends due to the bank’s failure. This effect will be examined
by measuring changes in share prices for affiliated companies, correlated
with price level and dividend. The ordinary least square regression is
expressed by:
54
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
where py stands for the general level of prices for the whole economy, divt
stands for general dividend series and pt represents the share prices for each
of the affiliated companies. The sample considered is 1861–75.
Table 2.8 does not reveal any significant figures regarding any influence
of the failure that are distinct from the effects of price level or dividends.
Dummy variables are introduced to concentrate the analysis on this specific
aspect. The table provides us with very low coefficients for both the GNP
deflator and dividends. If the figures produces by the second regression are
much higher, they remain insignificant as most of them are close to 0.
The regression equations which are going to be considered are:
55
Table 2.9 Relative impact of the bankruptcy of the Crédit Mobilier including a
dummy variable d1867
Table 2.10 Relative impact of the bankruptcy of the Crédit Mobilier including a dummy variable d1867/68
THE ROLE OF BANKS IN MONITORING FIRMS
To complete the argument the same regression can be carried out for d1867/
68
considering that d1867/68=1 in 1867 and 1868, 0 otherwise. The results are
given in Table 2.10.
Table 2.10 confirms the results obtained in the preceding test. Except for a
relatively small number of firms such as the Compagnie Immobilière and the
Compagnie Transatlantique, the bankruptcy did not lead to any great perturbation
that could not be explained by the price level (e.g. the GNP deflator coefficient)
or the financial market (e.g. the dividend coefficient). In what follows, these two
parameters will be avoided: we concentrate on the correlation between the
fluctuations of the share prices for affiliated companies and the bankruptcy, and
we evaluate the validity of an association with the Crédit Mobilier.
58
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
The aim of the test is to analyse the impact of the bankruptcy of the
Crédit Mobilier on the share fluctuations.
Specifically we intend to test the null hypothesis:
H0 β=0: ‘The bankruptcy of the Crédit Mobilier has no influence on the share
prices’ against H1 β<0
The results of the test are given in Tables 2.11 and 2.12.
These tables lead us to the following comments.
59
THE ROLE OF BANKS IN MONITORING FIRMS
two insurance companies are very similar to the those obtained for the
railways. The only comment that can be made concerns the size of the
regression coefficients. For the Omnibus Compagnie and for l’Union Life it
is of the same order as the railways. But as they are all positive the
conclusion is identical: H0 is not rejected by the data and there is no
evidence that the bankruptcy of the Crédit Mobilier had any influence on the
share prices.
The last point of our analysis will be to analyse the correlation between the
fluctuations of the share prices for affiliated companies and the variations in
asset prices for the Crédit Mobilier. With that aim in mind, monthly data
from 1866–8 are used to run two regressions. The first is expressed as:
The second has as its objective the evaluation of the interaction of the
change in asset price of the Crédit Mobilier on shares of affiliated
companies. The regression is given by:
All these tests are made separately for each year. A final regression will
consider the sample as whole for 1866–8. The results are presented as a set
of two tables representing the figures obtained for each of the two
regressions each. Standard errors are in brackets.
Some comments are worth making on these regressions. For the year
1866 the sign of the constant term for the Compagnie Immobilière changes
between the two regressions: in the first regression it is positive and in the
second one it is negative. What interpretation can we give to this
phenomenon? In the upper tables the t-statistics are uninterpretable as, under
the null hypothesis, the disturbances are log(p ) and are not independently
t
identically distributed or even probably stationary.
This set of results as a whole leads to the following interpretation:
except for the Compagnie Immobilière (where the β coefficient is strongly
positive, as shown by the first table of each year) and the Compagnie
Transatlantique, nothing serious happened when the Crédit Mobilier went
bankrupt. What can be said when considering the sample as a whole? The
regression figures are presented in Table 2.16 (standard errors are in
brackets).
Tables 2.13 to 2.16 reinforce the preceding arguments.
60
Table 2.13 Results for 1866
Table 2.14 Results for 1867
Table 2.15 Results for 1868
Table 2.16 Results for 1866–8
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
To complete the analysis, it will be interesting to extend the study to the firms
that were independent of the Crédit Mobilier. As in the preceding section,
65
THE ROLE OF BANKS IN MONITORING FIRMS
formal tests can be constructed to analyse the fluctuations of the share prices for
the companies which were not affiliated to the Crédit Mobilier. Let us consider,
then, the twelve steel companies and the Lenoir series to study the investment
constraint, if any, of these companies. A methodology identical to that followed
in the preceding section is used for the non-affiliated sample. In what follows,
the impact of the bankruptcy will be analysed, taking into account the general
effect of the bankruptcy on the financial market and the correlation between the
failure of the bank and the fluctuation of share prices for affiliated companies.
The aim of this section is to see if there was any marked influence on price
level and on the financial market due to the failure of the bank. This effect
will be examined in two stages:
• first, in the absolute, by evaluating the effect of price level and dividend
on share price for non-affiliated firms;
• second, by measuring the provoked modification in share prices for
non-affiliated companies in correlation with the change in price level
and dividend.
where py stands for the general level of prices for the whole economy, pt
represents the share prices for each of the affiliated companies and div is the
general dividend series. The sample considered is 1861–75.
The results are given by Table 2.17 (standard errors are in brackets).
As previously, Table 2.17 does not produce any significant figures
regarding any influence of the failure distinct from the effects of price level
and dividends (the coefficient obtained for the logarithm of py and dividends
being close to 0 or negative) dummy variables are introduced to concentrate
the analysis on this specific aspect. The regression equations which are
going to be considered are:
66
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
This test is done for each company separately and over the same sample,
that is 1861–75. The results are given by Table 2.18 (standard errors are in
brackets).
The statistical results given by Table 2.18 do not add any information to
the preceding tables. To complete the argument the same regression can be
done for d1867/68 considering that d1867/68=1 in 1867 and 1868, 0 otherwise.
The results are given by Table 2.19.
This comparison shows that there was no general increase in share prices in
1867 or 1868 in response to the Crédit Mobilier’s bankruptcy. For more than
half of the sample the coefficients for d1867 and d1867/68 are significant and
positive, while for the rest of the sample the same coefficients are slightly
negative and thus not really significant. On the other hand the argument
coefficients (ln(py) and ln(div) for the first regression, ln(pyt/pyt–1) and ln(divt/
div t–1 ) for the second one) are small and not significant. This shows
(unsurprisingly, given the results for affiliated companies) that the Crédit
Mobilier’s bankruptcy did not cause a general panic. In what follows, these
two parameters will be avoided: we concentrate on the correlation between the
fluctuations of the share prices for affiliated companies and the bankruptcy,
and we evaluate the validity of an association with the Crédit Mobilier.
67
Table 2.18 Relative impact of the bankruptcy of the Crédit Mobilier including a dummy variable d1867
Table 2.19 Relative impact of the bankruptcy of the Crédit Mobilier including a dummy variable d1867/68
THE ROLE OF BANKS IN MONITORING FIRMS
each company between 1862 and 1867. The equation of the regression
will be then:
ln (p p ) = c + βd
t / t–1 1867
70
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
(c) Conclusion
The Crédit Mobilier did not cause general distress (unsurprisingly, given the
results for affiliated firms). There is only weak evidence of a general boom
which could mash the effect of the Crédit Mobilier’s bankruptcy. The results
for affiliated and non-affiliated firms are very similar. This suggests that de
Long’s proposition does not apply to the Crédit Mobilier.
71
THE ROLE OF BANKS IN MONITORING FIRMS
For all the affiliated companies except the railways the values are given by:
–0.1125<β 0.097.
For all the affiliated companies except the railways the values are given by:
–0.1125<β<0.097.
As regards the standard error, affiliated companies as a whole show for d1867
a value smaller than the ones obtained for d1867/68 and close to 0. The same
consideration can be held for the non-affiliated companies.
What kind of interpretation can we give to these results? The railways
were best placed to resist bankruptcy for the following reasons:
72
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
As a whole, the affiliated companies did not suffer from the bankruptcy of
the Crédit Mobilier. The results obtained for the affiliated and non-affiliated
companies are very similar (coefficients are slightly higher for d1867/68, but
with a magnitude too small to be significant).
In conclusion, we can affirm that the bankruptcy of the bank had no
influence on either affiliated or non-affiliated companies. An association
with the bank does not seem to have increased the values of shares. This
expresses the lack of supervision of the Crédit Mobilier.
where CM=1 if the companies were affiliated to the Crédit Mobilier, CM=0
otherwise.
We run the regressions for the years 1866, 1867, 1868 and evaluate the
ratios:
ln p68 / p67
ln p68 / p66
The results, not significantly different for either ratio, are presented in
Tables 2.22 and 2.23.
Several comments can be made on Tables 2.22 and 2.23.
Table 2.22 Cross-section results excluding the Compagnie Immobilière
73
THE ROLE OF BANKS IN MONITORING FIRMS
• The figures obtained for the cross-section are not very different from
those resulting from the preceding OLS.
• The coefficient β in both regressions is close to zero. The only negative
sign appears when the Compagnie Immobilière is included.
• All t-statistics are insignificant.
74
THE CRÉDIT MOBILIER AND THE STOCK EXCHANGE
one, but this is, of course, subject to the ownership of the assets between the
shareholders. In particular, if one shareholder holds a huge amount of assets
of the company, the bank, in order to exert a supervisory role, must acquire a
majority, strictly larger than the external group. As regards the Crédit
Mobilier, if the participation in the firms was always major, nevertheless it did
not, typically, represent the largest proportion of the assets issued in the
companies. Hence the monitoring role of the bank was diminished, according
to this argument. However the strongest explanation of this lack of control
remains the bank’s risk-prone investment policy (e.g. the Compagnie
Immobilière).
Beyond this economic interpretation, a historical explanation can be
given. Around 1865 the Crédit Mobilier had lost some of its power. As early
as 1863, the Péreires requested permission to double the capital of the
Crédit Mobilier, but the government refused. With more and more of its
assets tied up in advances to the Société Immobilière, in 1866 the Mobilier
eventually persuaded the government to allow a doubling of its capital, in
return for modifications in its statutes restricting still further its freedom of
action. Naturally, the Péreires did not reveal to their stockholders the true
situation of the bank, but spoke vaguely and grandly of the need for new
resources to capitalise on the many opportunities for profitable operations
which daily came its way. In fact, of course, the Mobilier undertook no new
enterprises, but devoted all of the new capital to an attempt to salvage the
Société Immobilière.
Added to these difficulties, the Péreires faced a stockholders’ lawsuit
against the increase of the bank’s capital. In 1867, the Crédit Mobilier went
bankrupt. The first reactions of the governors of the Banque de France were
outrage and indignation. But when the financial and political implications
that the sudden failure of the Mobilier could provoke were pointed out, the
governors agreed, in effect, to permit a more gradual liquidation of the
enterprise. This constitutes one additional possible explanation of the lack of
fluctuations for the share prices of the companies with which the Crédit
Mobilier was involved.
75
3
CORPORATE INVESTMENT,
CASH FLOW AND FINANCIAL
CONSTRAINTS OF FIRMS
The case of the Crédit Mobilier
French industry with its tradition of self finance had few direct
links with the banking system…Whether or not this was so,
and if so whether it reflected the strength or weakness of
industrial enterprises, is still a subject for debate.
Cameron and Bovykin (1991:15)
3.1 INTRODUCTION
76
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
77
THE ROLE OF BANKS IN MONITORING FIRMS
There are several ways in which a bank and an industrial corporation can
interact. A company might conduct its daily business through a current
account with one or more banks. This was not the most common practice
with the Crédit Mobilier. In the case of a private company, a bank might
help to issue new equity or debt, for example bonds. A bank might take a
direct stake in the company, or it appoint representatives to sit on the
company’s board of the directors.
This last arrangement was the most frequent for the Péreires’
establishment. In every company financed by the Crédit Mobilier, a member
of its board of directors was represented in the shareholders’ assembly.5 The
question of the advances to companies which were granted by the Péreires
emphasises the direct participation of the bank.
Lending, issuing shares, direct investment and supervisory board
members helped the bank to monitor and in many cases influence decision-
making in the industrial enterprise. Nevertheless, Jeidels (1905) carefully
distinguishes between monitoring, in the sense of information-gathering, and
direct influence on a corporation’s business decisions. Although collecting
information seems to have been likely in the Crédit Mobilier’s case, any
kind of direct influence over the decisions of management seems improbable
(cf. the empirical results in Chapter 2). However, generally speaking, placing
bank directors or members of the bank onto the board of corporations was
common practice for the Péreire brothers and could have been the most
effective means of influencing managerial decisions of affiliated
corporations. In Chapter 2 we tried to analyse whether this supervision was
effective in the Crédit Mobilier’s case. We intend to reinforce this argument
by examining the investment policy of the bank.
The aim of this section is then to analyse the relationship between
investment and cash flows between 1860 and 1880. Two samples are
78
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
The principal source for data underlying the statistical analysis comes
from the General Assembly’s reports and some articles and books such
as those by Jean Denuc (1934) and Bertrand Gille (1968). In his book,
Bertrand Gille has reproduced a great number of company balance
sheets, particularly steel companies; these figures make it possible to
deduce cash flows, investment and book values. Jean Denuc provides
values for other companies; the railways companies were reported
separately. The different indices Denuc gives for all series enable us to
deduce the book value and the cash flows for the different companies.
The investment factor was given by an article issued by the Institut
National de la Statistique which recapitulated the economic situation of
France from 1850 until 1914. Here, we want to evaluate the correlation
between cash flows CF, common equity Q and investment I. The
regression specification was taken directly from Hoshi, Kashyap, and
Scharfstein (1989, 1991) and was also used in Ramirez (1992). Table
3.1 lists the companies included in the sample taken into consideration
and records their affiliation with the Crédit Mobilier. As can be seen,
railways companies represent the major part of the Crédit Mobilier’s
investments.
We can note also that all companies were affiliated with either the Crédit
Mobilier or the Crédit Lyonnais, both as powerful banks over the period.
When the banks offered credit for business firms at this time, industrial
enterprises were rarely involved because they had been warned by
experience. Significant amounts of capital were provided by merchant banks
especially to railways, shipping companies and urban transport projects, but
industrial enterprises received little. As far as we are concerned with public
utilities, the bank’s affiliation was omnipresent. This suggests that it may be
difficult to deduce the effect of affiliation with a big bank by comparing
79
THE ROLE OF BANKS IN MONITORING FIRMS
80
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
and compare these balance sheets. In order to avoid any ambiguity, this
parameter will be called Cb, which should represent for the French case the
gross value of shares.
All variables, the market prices of common equity, the cash flows CF and
the investment I, are normalised by the stock of depreciable assets at the
beginning of the period. First, some summary statistics comparing Crédit
Mobilier firms and non-Crédit Mobilier firms are presented. As the Crédit
Mobilier went bankrupt in 1867, the mean is calculated in constant terms in
regression, taking three samples:
Table 3.2 Summary statistics comparing Crédit Mobilier and non-Crédit Mobilier
firms*
81
THE ROLE OF BANKS IN MONITORING FIRMS
Note: Averages are calculated for all firms and all years. Standard errors are in
parentheses. Capital (Cb) and investment (I) are for depreciable assets. Sources and
definitions of variables are described in the text. The sample for the Crédit Mobilier
firms is constituted of seven firms, for non-affiliated firms eight.
century was characterised by the odd, scrappy profit calculation and balance sheet
which contained assets valued in a wide variety of ways. In fact the main aim of
the published balance sheet was to provide creditors and shareholders with a
statement of company solvency and to reassure them that dividends had not been
paid out of capital. There was no requirement for the publication of a profit
statement, nor were there any minimum disclosure rules regarding the content of
the balance sheet. During the nineteenth century the calculation of profit became
the first stage in the preparation of the final accounts. The new approach to profit
was the use of profits as the basis for dividend declarations. This adoption of the
matching concept reduced the scope for management to manipulate profit; with
valuations, there is a need not only to make estimates, but also to decide on the
appropriate method of valuation from the various options available, including
present value, selling price and replacement cost. This last parameter is of such
importance that we should refer to James Tobin’s Q theory of investment.6
This can be summarised as follows. The rate of investment—the speed at
which investors wish to increase the capital stock—should be related, if to
anything, to Q the value of capital relative to its replacement cost.
Over the period, replacement accounting delayed the recognition of
capital consumption until expenditures were made for renewals.
Furthermore, since periodical renewals would require substantial sums, there
was a bias against renewals as costs because replacement reserves were not
generally maintained. On the one hand, a firm might not have been able to
charge substantial renewals to expenses without creating a deficit, while on
the other hand, without sufficient internal funds available, it would be
necessary to finance replacement expenditures with sources external to the
firm. This discussion suggests that the application of replacement accounting
tends to create serious liquidity problems and that renewals might have been
deferred or treated as additions in many cases.
82
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
It/Kt = α0 + α1 CFt/Cbt + α2 Q
where:
The main results are presented in Tables 3.3, 3.4 and 3.5. The evidence
presented in these three tables suggests that free cash flow was more
important in both magnitude and statistical significance among independent
firms. That means that free cash flow was an important determinant of
investment expenditures among the independent mining and steel companies.
This is also evidenced by the fact that the overall pattern of the regression
for affiliated companies is significantly different from that of the non-
affiliated companies. That is, we cannot prove that the explanatory variables
are explained by different investment expenditures among independent and
affiliated companies.
Included in these tables is an interaction dummy which takes the value 1
times cash flow when the corporation is affiliated to the Crédit Mobilier.
This variable is introduced to estimate the difference and statistical
importance of the cash flow coefficient between affiliated and non-affiliated
companies.
83
THE ROLE OF BANKS IN MONITORING FIRMS
84
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
85
THE ROLE OF BANKS IN MONITORING FIRMS
Conclusion
The results obtained for France in the Crédit Mobilier case are very different
from those presented for the United States and Japan by de Long and
Ramirez, or for Germany by Marco Becht. Substantially, de Long (1989)
asserts that the monitoring role of the Morgan Corporation was, in general,
successfully achieved. The Pujo Committee of 1913 agreed that the House
of Morgan had sufficient resources and influence in the financial world to
render it the most powerful and active of the few investment houses. The
econometric work of Ramirez (1992) suggests that association with
J.P.Morgan relieved a credit constraint.
The evaluation of Q and its fluctuations is not completely different from
the results obtained by de Long and Ramirez. However, the participation of
the bank in the firms’ projects was so non-uniform that a direct association
with the Crédit Mobilier did not always contribute to an increase of Q for
the affiliated firms (Table 3.5). It seems on the contrary that the bankruptcy
of the Crédit Mobilier removed a liquidity constraint for affiliated firms. The
comparison of the results obtained for 1861–7 on the one side and for
1868–75 on the other side indicates that, after the failure, the affiliated firms
did not suffer from liquidity constraint as they had before.
As a whole, our results seem to indicate that association with the Crédit
Mobilier created liquidity problems. The next section intends thus to isolate
the bankruptcy effect in order to evaluate more precisely the profitability of
an association with the Crédit Mobilier (if any) on Q.
86
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
The regression equations are evaluated by introducing year dummy and firm
dummy variables. In other words, we determine investment for every year
except for one, separately, and for every firm except one, separately. As in
the preceding test we consider three samples:
I/Cb = α0 + α1 Q + γ.CF/Cb
where:
87
THE ROLE OF BANKS IN MONITORING FIRMS
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
in brackets: affiliated firms with the Compagnie Immobilière 7, independent companies
8, all firms 15.
The regression equations are evaluated by introducing year dummy and firm
dummy variables. In other words, we determine investment for every year
except for one separately and for every firm except one separately.
Let us consider first the Crédit Mobilier firms and non-affiliated
companies. Tables 3.6, 3.7 and 3.8 present the main results.
Some general remarks can be made regarding these tables:
88
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
Table 3.7 Fixed effect model for 1861–7*
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
in brackets: affiliated firms with the Compagnie Immobilière 7, independent companies
8, all firms 15.
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
in brackets: affiliated firms with the Compagnie Immobilière 7, independent companies
8, all firms 14.
Note: *The dependent variable is investment relative to capital stock at the beginning of
the period. The regressions include year dummies and firm dummies. Standard errors in
brackets. The number of companies is 15 for 1861–67 and 1861–75, and 14 for 1868–75.
89
THE ROLE OF BANKS IN MONITORING FIRMS
French case. The Crédit Mobilier did not act as a universal bank throughout
the whole period. Because of asymmetric information, non-affiliated
companies face much higher costs of external financing than did affiliated
ones. However, this cost was partly compensated by the risky behaviour of
the bank towards investment.
How can we explain these differences between France and Germany?
A series of eminent economic historians 7 had held that the major
innovation in industrial financing in the nineteenth century was the
Crédit Mobilier. Gerschenkron’s theory was elaborated into a model, a
backwardness, in which banks, depending on how backward a country
was, substituted liquidity for entrepreneurship. This had been the driving
force in the industrial revolution in the United Kingdom, for example.
But evolution proceeded rather differently in other countries. German
banks took the example of the Crédit Mobilier and developed it to the
utmost. Apart from Hamburg and Frankfurt, which clung longer to
commercial banking, the German banks went in for close relations with
industrial and mining firms.
Theoretically, the Crédit Mobilier’s case illustrates perfectly the following
dilemma. Adherents of rational expectations and efficient markets tend to
believe that markets always work. Institutionalists, of course, take the
opposite view, that institutions determine historical outcomes. Between the
two positions, there is room for an eclectic view that markets mostly work
but occasionally break down, and that sometimes institutions adapt to
underlying changes in demand and supply conditions and sometimes they do
not. The Crédit Mobilier was not capable of adapting to the change.
(d) Conclusion
The main premise of this section is that, unlike the Morgan company, 8
the Crédit Mobilier was not fulfilling an effective monitoring role but
only provided liquidity to firms whenever necessary. This bank does not
seem to have alleviated the impact of the imperfections of the capital
market.
To strengthen the argument, we consider two extensions of this work in
the following chapter. The first step will be to include in the book value a
depreciation rate δ, equal to 5 per cent, in order to conform more to the Q
theory. All details for the description of the parameters will be given in
Chapter 4. Some additional value for δ will be considered to evaluate the
robustness of the test. The next step will be to regress investment, cash flow
and Q on the basis of their real value, that is by dividing the parameters by
the general level of prices in France.
90
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
91
THE ROLE OF BANKS IN MONITORING FIRMS
Note: *Averages are calculated for all firms and all years. Standard errors are in
parentheses. The sample for the Crédit Mobilier firms is constituted of 7 firms over
1861–7 and 1861–75, and 6 firms over 1868–75. The number of affiliated firms is 8.
which would accumulate to the desired sum by the time replacement became
necessary was another. This explains the flat series for Cb. How can we
transform it, and what is the definition of the new parameter?
Let Cbk1861 represent the first element of the series which is defined as:
Cbk1861 = I1861/δ
I/Cb = a + a Q + ?.CF/Cb
1 0 1 1 1
where:
92
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
Note: Standard errors are in brackets. *CF.CM represents CF times Crédit Mobilier
affiliation: affiliated firms 7, independent companies 8, all firms 15.
In the following section we present the results obtained and compare them
to those of the preceding section. As before, we consider three samples:
the whole sample 1860–75, the sample prior to the bankruptcy i.e. 1860–7,
and the years after the bankruptcy. The results are given in Tables 3.11,
3.12 and 3.13.
If we compare the results obtained here with those of the preceding
section, we can make the following comments. Let us first consider the
whole sample 1861–75. The coefficient for CF/Cb and Q are not very
1 1
different for affiliated and independent firms. The association with the
Crédit Mobilier, even if it was effective (largest value for the dummy
coefficient), was not important enough to cause a difference between the
two sets of companies.
As regards the sample 1861–7, a comparison with the results obtained by
Becht and Ramirez (1992:28) will be interesting. Let us recall them briefly.
Tables 3.14 and 3.15 only give us the coefficients obtained by running a
regression similar to the one presented in this chapter. We deliberately
choose not to report the standard errors, the aim being to analyse the
analogies and the differences between the two countries.
93
THE ROLE OF BANKS IN MONITORING FIRMS
Table 3.12 Investment regression over 1861–7
Note: Standard errors are in brackets. *CF.CM represents CF times Crédit Mobilier
affiliation: affiliated firms 7, independent companies 8, all firms 15.
Note: Standard errors are in brackets. *CF.CM represents CF times Crédit Mobilier
affiliation: affiliated firms 6, independent firms 8.
If we compare these figures with the ones we obtain, we remark that the
coefficients of the cash flows and the common equity are in exactly the
reverse order for the Crédit Mobilier:
• CF/Cb is close to 1 for the affiliated companies and close to zero for
1
independent firms, which means that the affiliated companies were
94
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
Table 3.14 Results for affiliated companies as presented in Becht and de Long’s
paper
de Long’s Paper
95
THE ROLE OF BANKS IN MONITORING FIRMS
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
are in brackets. Affiliated firms with the Compagnie Immobilière 7, independent
companies 8, all firms 15.
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
are in brackets. Affiliated firms with the Compagnie Immobilière 7, independent
companies 8, all firms 15.
where α0=intercept, αi=(N–1) firms specific fixed effects, and γt=(T–1) year
dummies and e is a residual term.
The results from these estimations are given by Tables 3.16, 3.17, 3.18
and 3.19.
If we analyse the coefficient we obtain for the cash flows, we notice that
the Crédit Mobilier firms were more constrained before the bankruptcy (CF/
Cb =0.112 for Crédit Mobilier firm over 1861–7) than after (where CF/
1
Cb =–0.103). This is the opposite of what is predicted by theory. The
1
collapse of the Crédit Mobilier should have created liquidity problems for
affiliated firms. The most remarkable result of Tables 3.16 and 3.17 is the
96
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include yearly dummies and firm dummies. Standard
errors are in brackets. Affiliated firms 6, independent companies 8.
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
are in brackets. All firms with the Compagnie Immobilière 15.
97
THE ROLE OF BANKS IN MONITORING FIRMS
(d) Conclusion
All the empirical work in this section is based on the assumption that the
depreciation rate is equal to 5 per cent. This choice can be seen as arbitrary,
but leads to a slightly increasing series for Q. To complete the analysis
presented here we introduce the modification implied in the data set if we
consider a depreciation rate d greater and smaller than 5 per cent. Appendix F
presents the figures obtained for the coefficient Q when introducing a
Table 3.20 Investment regression equation for affiliated companies and independent
firms over the whole sample*
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
are in brackets. Affiliated companies with the Compagnie Immobilière 7, independent
firms 8.
98
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
Table 3.21 Investment regression equation for affiliated companies and independent
firms before the bankruptcy*
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
are in brackets. Affiliated companies with the Compagnie Immobilière 7, independent
firms 8.
depreciation rate of 3 per cent and 7 per cent. In what follows, the new
variables for Q and Cb are noted Q2 and Cb2 when δ=3 per cent, Q3 and Cb3
when δ=7 per cent.
The aim of this section is then to measure the modification induced by a
small difference in the depreciation when analysing the correlation between
investment, cash flows and common equity. For that purpose we concentrate
the econometrical analysis on the fixed effect model. The previous tests
presented in section 3.2(a) are not going to be considered, in order to
alleviate a fastidious examination of a large number of tables. The
regression we run considers the set of affiliated companies on the one side
and the set of independent companies on the other side over three samples:
the whole sample 1861–75, the situation before the bankruptcy 1861–7, the
event after the bankruptcy. We then evaluate the correlation between
investment and cash flows for the whole set of companies over the same
samples. The results presented in Tables 3.20, 3.21 and 3.22 successively
report the figures obtained for the two values of depreciation rates in order
to facilitate their interpretation.
99
THE ROLE OF BANKS IN MONITORING FIRMS
Table 3.22 Investment regression equation for affiliated companies and independent
firms after the bankruptcy*
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
are in brackets. Affiliated companies 6, independent firms 8.
Note: *CF/Cb1 stands for cash flows including a depreciation rate of 5%; CF/Cb2
stands for cash flows including a depreciation rate of 3%; CF/Cb 3 stands for cash
flows including a depreciation rate of 7%.
Some general comments can be made on this first set of results. In order
to summarise the arguments we construct a small table (Table 3.23)
reporting all the figures we obtain for the coefficient on cash flows.
100
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
As can be noted, the coefficients for the cash flows are very similar for
affiliated companies when the Compagnie Immobilière is excluded. whatever
level of deprecation rate is considered. The same argument can hold for the
independent companies.
As regards the coefficient on Q, the preceding tables exhibit very strong
similarities for every set of companies (affiliated companies with or without
the Compagnie Immobilière, as well as independent companies). The
constant term always has the same sign and the same magnitude. The same
conclusions follow for R2.
Some additional comments can be made on the fluctuations of the t-statistics
as regards the robustness of this test. In what follows we concentrate on the
whole sample (1861–75) and on the years before the bankruptcy (1861–7)
because the rest of our results exhibit no big change for the t-statistics. More
specifically we consider the set of companies excluding the Compagnie
Immobilière. The analysis is summarised by two tables (Tables 3.24 and 3.25)
which reproduce the coefficient on CF/Cb in order to appreciate the degree of
i
liquidity constraints for affiliated companies over these two period.
First, it is remarkable that the introduction or the exclusion of the Compagnie
Immobilière should have an influence on the interpretation of the liquidity
constraints that the affiliated firms had to suffer during these two periods. The
Table 3.24 Sample 1861–75 robustness results for Crédit Mobilier firms without the
Compagnie Immobilière
Table 3.25 Sample 1861–7 robustness results for Crédit Mobilier firms without the
Compagnie Immobilière
101
THE ROLE OF BANKS IN MONITORING FIRMS
Table 3.26 Investment regression equation for all companies over the different
samples
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include yearly dummies and firm dummies. Standard
errors are in brackets. Affiliated firms with the Compagnie Immobilière 7, independent
companies 8, all firms 15.
figures reported in these tables show that the coefficient on cash flows is more or
less significant according to the level of depreciation rate included in the
regression. As this situation does not represent the norm over the entire statistical
work presented here, we can say the results are robust: the sign and the amplitude
of the point estimates have not changed when the depreciation rate varies.
Let us now consider the whole set of firms for each sample. The results
are presented in Table 3.26. In order to compare the figures obtained for the
three values of d we construct two tables (Tables 3.27 and 3.28) which
summarise the results obtained for Q and CF/Cb . Some general comments
i i
will be made for the dummy coefficient and the R2 parameter. Let us first
consider the coefficient obtained for CF/Cb .
i
The values reported in Table 3.27 exhibit strong similarities for the three
values of d, except in one case, CF/Cb for all companies 1861–7 when the
3
102
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
study the parameter Q and to examine if the same particularity happens over
the same sample.
As regards Table 3.28 it seems that the similarities for this coefficient are
even stronger. If we now consider the dummy variable, we see that it always
has the same sign and the same magnitude. The same line of argument can
be made for the constant term.
For all firms, R2 is included between:
0.647 ≤ R2 ≤ 0.995
which means that the correlation between the variables is quite high.
As before, we add some indication as regards the variation of the t-
statistics, concentrating on the only sensible sample, 1861–7. Two
coefficients are taken into consideration: the cash flow and the dummy
variable, reported in Table 3.29.
The most surprising element of this table is the small value for the t-
statistics obtained for the dummy variable when the depreciation rate is
equal to 5 per cent. If this number is compared to the whole set of figures
reported in the table, it seems that this situation represents a marginal case
and does not alter the validity of our results.
103
THE ROLE OF BANKS IN MONITORING FIRMS
Table 3.29 T-statistics for sample 1861–7 as regards the cash flow parameters
(a) Conclusion
The appendix describes in detail the series retained for Pt and the method
used to compute Q. Let us present some descriptive statistics to analyse the
modifications produced by this alteration of the data.
If we compare these figures with those obtained previously we note that
they do not conform to what could be expected.9 Taking them as correct the
point is then to evaluate the impact of this modification on the investment
regression.
The selected regression specification takes the following form:
104
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
Table 3.30 Summary statistics comparing Crédit Mobilier and non-Crédit Mobilier
firms when parameters are expressed in real terms*
Note: *Averages are calculated for all firms and all year. Standard errors are in
parentheses. Affiliated firms with the Compagnie Immobilière 7, independent firms 8.
Note: *The dependent variable is investment relative to capital stock at the beginning
of the period. The regressions include year dummies and firm dummies. Standard errors
are in brackets. Affiliated companies with the Compagnie Immobliere 7, independent
firms 8.
where: α0=intercept, αi=(N–1) firms specific fixed effects and γt=(T–1) year
dummies.
The results from these estimations are given by Tables 3.31, 3.32, 3.33 and 3.34.
Some comments can be made on these four tables. Let us first consider
what happens over the whole sample and for 1861–7. The first observation
is that the coefficient on Q is larger for affiliated companies and the second
2
that the results obtained when the Compagnie Immobilière is part of the
sample are much different from the ones where it is excluded.
As regards the independent companies, the coefficients are not well
estimated. This difference appears more pronounced than in the preceding
test. One reasonable explanation could be the lack of data: missing values
are introduced for 1861–4.
105
THE ROLE OF BANKS IN MONITORING FIRMS
Table 3.32 Investment regression equation for 1861–7*
Note: *The dependent variable is investment relative to capital stock at the beginning of
the period. The regressions include year dummies and firm dummies. Standard errors are
in brackets. Affiliated companies with the Compagnie Immobliere 7, independent firms 8.
Note: *The dependent variable is investment relative to capital stock at the beginning of
the period. The regressions include year dummies and firm dummies. Standard errors are
in brackets. Affiliated companies 6, independent firms 8.
Note: *The dependent variable is investment relative to capital stock at the beginning of
the period. The regressions include year dummies and firm dummies. Standard errors are
in brackets. All firms with the Compagnie Immobilière 15.
106
CORPORATE INVESTMENT, CASH FLOW AND FINANCE
107
THE ROLE OF BANKS IN MONITORING FIRMS
Mobilier. This again suggests that the failure of the Crédit Mobilier relaxed
liquidity constraints for affiliated firms.
All these figures prove that the Crédit Mobilier’s firms were more liquidity
constrained before the bankruptcy than after. The bank never managed to
supervise firms. As long as stock values were appreciating, the Crédit
Mobilier found it easy to participate in the stock market while diversifying
its activities. The bank participated in railway buildings in Europe,
international banking and French real estate. This multifaceted activity was
in part due to the Péreires’ very broad view of the goals of the Crédit
Mobilier, but it was also a response to the booming international capital
market that occurred after mid-1850.
The extraordinary degree of activity of the Crédit Mobilier meant that it
operated with few reserves, especially given that it distributed a substantial
part of its profit for nearly a decade. The absence of reserves made it
impossible for the bank to survive the economic slowdown that followed
1865 and relieved credit constraints for affiliated firms.
The next step of our argument will be to discuss the place of the Crédit
Mobilier in the banking system, in order to explain the minor effect on
companies when the Crédit Mobilier collapsed. We pay attention in our
analysis to the differences between France, Germany and the United States
over the same period.
108
4
4.1 Introduction
109
THE ROLE OF BANKS IN MONITORING FIRMS
110
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
If the Crédit Mobilier were truly a great bank, its bankruptcy would have
provoked financial difficulties not only for the enterprises with which it was
involved but also for the other banks on the market.
111
THE ROLE OF BANKS IN MONITORING FIRMS
112
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
113
THE ROLE OF BANKS IN MONITORING FIRMS
114
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
115
THE ROLE OF BANKS IN MONITORING FIRMS
(c) Conclusion
The results for the Crédit Mobilier’s case are different from the
German Goßbanken or the J.P.Morgan Company. Becht and Ramirez
conclude that ‘mining and steel companies affiliated with one of
Imperial Germany’s “Great Banks” were not liquidity constrained’.
Their paper provides evidence suggesting that this limit can be very
high. The power of the Großbanken seems to have helped the
development of the mining industry by alleviating the impact of capital
market imperfections.
The Crédit Mobilier’s case, however, proves the fragility of banks’
power; it shows that the supervisory role can be very limited and
dispensable. The Crédit Mobilier was vulnerable to speculative
movements. It is possible that the hostility of famous financiers like the
Rothschilds contributed to its continuously decreasing power in financial
circles.
116
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
To begin our thesis, let us give some of the background to the banking
sector in the nineteenth century. During the period of this study there were a
number of sounder financial institutions than the Crédit Mobilier, which had
an enormously swollen industrial portfolio greatly exceeding its capital, and
which was dependent on favourable developments on the stock exchange to
support the continuation of its activities. Universal banks such as J.P.Morgan
and the Großbanken established the closest possible relations with their
client industrial enterprises. Universal banks accompanied an industrial
enterprise from the cradle to the grave, from establishment to liquidation
through all the vicissitudes of their existence. Through the device of
formally short-term, but in reality long-term current account credits and
from the development of the institutions of the supervisory boards to the
position of the most powerful organs within corporate organisations, the
banks acquired a formidable degree of power over industrial enterprises,
which extended far beyond the sphere of financial control into that of
entrepreneurial and managerial decisions. This argument is strongly
supported by evidence presented by de Long and Becht, Ramirez and Becht
and Ramirez.
Association with the Morgan company or with one of the Großbanken
appears to have added value to enterprises. These were banks that took an
active part in the decisions related to choosing a project, in order to defend
their reputation in the financial markets. In the Crédit Mobilier’s case this
argument is problematic, for one essential reason: as noted above, the bank
extended its role to that of a shareholder (and often the principal
shareholders of the firm). Therefore it suffered the same risks as an ordinary
shareholder.
The answer to the question: ‘Is it preferable for banks and firms to be
associated or not?’ is neither always yes nor always no. For Aoki (1988) the
bank can be a monitoring agent. Two essential reasons can be given to
support this proposition:
1 The firm can raise investment funds by borrowing from the bank at a
low cost, and does not have to rely on equity financing. Therefore, its
strategic business decisions (corporate policy) can be made in the
interest of individual stockholders.
2 Individual stockholders of the firm are underprivileged in the sharing of
economic returns to the firm, as they receive only a very small fraction
of profit as dividend.
117
THE ROLE OF BANKS IN MONITORING FIRMS
118
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
The Crédit Mobilier had attempted a new style of banking, but never
managed to apply it correctly: it still stuck to the old style, which consisted
of holding 51 per cent of the assets of a company in order to participate
actively on its financing. The companies affiliated to the Crédit Mobilier
were joint stock companies managed like limited partnerships, which
corresponded to the French tradition.
Concerning the advantage for a firm of an association with a specific
bank, Mayer (1988) suggests that there exist systematic differences in
performance between the financial systems where banks play a primary
role and those where the banks are not active in the same way. These
differences must be explained by the mechanisms that each system
provides to suppress the moral hazard in relations between entrepreneurs
and financiers. More specifically, Mayer affirms that the systems
proposed by the German or Japanese banks enable the firms and the
banks to sustain long-term contracts which in turn enable them to extract
the surplus associated with these contracts and to reduce moral hazard.
The answer to the question of the (dis)advantage of bank-firm
association will substantially depend on:
• the proportion of assets held by the bank: are the decisions taken by
the shareholders’ assembly the expression of the majority or, as in the
Péreires’ case, the will of the founders? The results of our tests prove
that the second alternative is more realistic in the Crédit Mobilier’s
case. In fact this argument leads us to analyse the position of
shareholders when a bank holds a large amount of assets in a firm.
Considering the Crédit Mobilier’s case, it seems that the role they
played was limited. We will discuss this idea in the next section when
comparing the French case to the German one;
• the expectations of financiers on the possible profitability of the firm
(e.g. confidence on the potential revenue for the bank in the future); this
reflects essentially the bankers’ point of view, whose aim is to make
profit;
• the power the bank exerts on the financial market (e.g. capacity to solve
the capital market imperfections). This last point corresponds essentially
to the situation of the German Großbanken, in which case an association
was profitable on average: for example adding value to the enterprise’s
shares.
Conclusion
It is relevant to point to the stress laid by Saint Simon and his followers
upon industrialisation and the great task they had assigned to banks as an
instrument for organisation and development of the economy. This, no
doubt, appealed greatly to the creators of the Crédit Mobilier, who liked to
119
THE ROLE OF BANKS IN MONITORING FIRMS
120
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
The cash currently held in our bank was in great part hidden in the
drawers, without any interest for those who kept it, without any utility
for French society.5
The aim of the Crédit Lyonnais was quite different from the Crédit Mobilier.
In 1854, Isaac Péreire had declared that the Crédit Mobilier was ‘a limited
partnership’ for industries, 6 a lending bank and a bank of issue. The
ambitions of this bank were wider than those of the Crédit Lyonnais. This
diversification made the Crédit Mobilier powerful, but was also at the origin
of its difficulties. Specification of its different activities will enable us to
understand the functioning of the establishment.
Table 4.1, after Bouvier (1968), illustrates this latter point. Bouvier asserts
that during its early years, the Crédit Lyonnais was a kind of readjusted
Crédit Mobilier, less ambitious, more prudent and more solid.
If we study Table 4.1, we note that the ‘big deals’ only exceptionally
121
THE ROLE OF BANKS IN MONITORING FIRMS
represented more than half of the total transactions of the bank. This
enabled the bank to avoid liquidity problems during the French financial
crises of 1870 and 1875. The Crédit Mobilier, on the other hand, was
closer to what Sayous (1907) calls a financial firm. Talking of these
firms he said:
But they (the financial firms) generally play a more active role: they
hold a block of shares, prepare the market or wait for any favourable
circumstance and pass the assets to the public; they themselves
transform the private enterprises into limited companies from which
they later issue certificates. They go as far as to create industries or
to organise business using an idea which was suggested to them.
They are then the brokers, the traders of this modern merchandise
called stocks: they facilitate and even organise the contact between
savings and those who need it.7
122
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
(c) Conclusion
The points outlined above have a bearing on the validity of the econometric
results set out earlier. All the statistical analysis has to be interpreted in the
123
THE ROLE OF BANKS IN MONITORING FIRMS
The next point we are going to consider is the divergence of the results
obtained in the French and the German cases. As noted at the beginning of
this chapter, the results obtained for Germany and France are very different.
The aim here is to give some interpretative reasons for this difference. More
specifically, we examine if the type of participation of the Großbanken in
firms followed the same model as the Crédit Mobilier.
The financial commitment of every ‘mixed bank’ consists in the
promotion and financing of industrial firms (short- and long-term financing:
the long-term corresponding to the renewal of short-term credit and
advances). The German financiers were real entrepreneurs, facilitating access
124
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
125
THE ROLE OF BANKS IN MONITORING FIRMS
If all these arguments seem to stress the prudent and rational attitudes
as regards credit policy, one point should be mentioned which concerns
the question of advances to companies. Sayous (1899) affirms that the
economic development of Germany, in establishing a strict distinction
between issuing banks and ordinary banks had, as a consequence,
facilitated the credit and financial operations; the relationship between
banks and industry became stronger and stronger as the activity
increased. However, the politics as regards advances was not different
at all. Simple loans based on certificates and assets did not play a
crucial role under the classical form in the relations between trade and
industry; more important were the credits granted on current accounts.
If the name differed in Germany, there is no doubt that the results were
identical. Both are advances granted to industries without any special
guarantee.
The success of the Großbanken may simply have been due to the fact that
they developed around 1870, twenty years after the mistake of the first
Crédit Mobilier.
126
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
(a) Conclusion
The eformer situation, where the bank held the main part of the assets in
company, would have been impossible in the German case. For the
Großbanken this would have signified difficulties for the companies in
negotiating their titles on the equity market. Such a ‘signal’ would have
induced non-participation for a German bank (Whale 1930). This distinction
is very useful with regard to the power relationships between the firms and
the Crédit Mobilier.
For the first subsample, power was derived from the leading position the
Crédit Mobilier held by virtue of the number of shares belonging to the
Péreires. However, one surprising point is that the other shareholders were
strong enough to force the Péreires to resign after 1867. How can this
conundrum be explained? For this subsample (excluding the Compagnie
Immobilière), the Crédit Mobilier served the firms by providing sufficient
liquidity for the firms to grow during the period of French industrialisation.
In terms of the principal-agent theory, it seemed that the bank—in
appearance the principal—in reality was an agent which served the whole
firm (that is, the other block of shareholders). This perhaps helps to explain
the powerful situation the firms were in when the bank failed.
For the second subsample, the roles were well defined: the bank (the
principal) chose to grant credit to the firms because it expected thereby
to earn some profit. However, something is odd in this case: the firms
did not suffer from the bankruptcy. If an association with the Crédit
Mobilier had been a signal of reliability, the shares of firms should have
declined in value when the bank went bankrupty and thus lost its
reputation for sound financial judgement. This suggests that the Crédit
Mobilier never had such a reputation.
127
THE ROLE OF BANKS IN MONITORING FIRMS
128
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
towards the policy of the bank. We can discuss whether this high
dividend policy did not correspond to a partial liquidation of the
establishment. Perhaps, since shareholders received a high enough
dividend for four years, they considered that, even if the worst
happened, they would still gain.
This quotation points out the enthusiasm of the shareholders for a firm
which had given them huge dividends and managed in the past to endure all
speculative movements.
According to the traditional theory of the firm, this attitude was
irrational. The firm must, generally speaking, be considered as a profit-
maximising entity, that is:
The case of the Crédit Mobilier contradicts these two propositions. The
Péreires did not act in the shareholders’ interests, but in their own interests.
As they held the major part of the capital of the bank, the power of the
other shareholders was very limited. Hence the control policy inside the
bank was not effective. This constitutes a good example of the difficulties in
supervising companies when financiers are at the same time decision-makers
in the productive process.
4.6 CONCLUSION
129
THE ROLE OF BANKS IN MONITORING FIRMS
In the Crédit Mobilier’s case nothing like this was done. The only task the
Péreires accomplished was to provide liquidity and to try to realise an
immediate profit, without really taking care of the solvency of their own
establishment.
Another issue concerns whether the bank-firm relationship consists of a
domination by the banker, or a more flexible contract. As regards the Crédit
Mobilier, the policy of placing a member of the bank on the board of
directors in each company demonstrates the intention of the banker to place
himself in a leadership position. But the results that followed do not bear
out this assumption.
Close relationships between banks and industry developed everywhere
basically on the same lines during the nineteenth century. However, the evolution
of these relationships proceeded rather differently in different countries.
Geographically, my illustrative evidence is limited to Germany and the United
States. The whole set of results exhibited by the Crédit Mobilier’s case contrasts
with those obtained for the German Großbanken or the J.P.Morgan company.
Considering then the existing theory concerning the investment policy of
firms regularly associated with a bank, it is interesting to establish their
predictive power. Our results have two specifics:
130
THE SUPERVISORY ROLE OF THE CRÉDIT MOBILIER
They prove anyway that, contrary to the suggestion of Bernanke and Gertler
(1989), capital market imperfections do not always contribute to excessive
output fluctuations. For the pre-WWI period both assumptions, the scarcity
of capital (illustrated in France by an underdeveloped capital market
substituted by powerful financiers) and the decisive role of banks in
furthering industrialisation were tested. The main findings with regard to the
Crédit Mobilier serve to intensify the debate on the assumption of the
banks’ supremacy over industry.
131
5
GENERAL CONCLUSIONS
This book has considered the supervisory role between 1852 and 1880 of
the Crédit Mobilier, a bank which went bankrupt in 1867 and was
reorganised subsequently.
Economic historians have often studied the prominent role played by the
Crédit Mobilier in industrial financing in the nineteenth century.
Gerschenkron elaborated a model of backwardness, in which first banks,
then government, depending on how backward a country was, substituted
themselves for the entrepreneurship that had been the driving force in the
industrialisation of the United Kingdom, called ‘the first industrial nation’
by Peter Mathias. Cameron explained in detail how the Crédit Mobilier and
its imitators were responsible for the economic development not only of
France but of Continental Europe as a whole.
In order to evaluate the impact of monitoring on this bank, two
approaches have been adopted. First, we have studied the impact of the
Crédit Mobilier’s bankruptcy on share prices for affiliated and non-affiliated
companies to ascertain if—as in de Long’s study of J.P.Morgan —an
association with the Crédit Mobilier added value to the firms with which the
bank was involved. The sample we have constructed rejects such a
hypothesis. No significant effect of the bank’s bankruptcy could be detected
either on the economy as a whole or on the particular sample of twenty five
firms, half of which were involved with the Péreire brothers, the others
being totally independent of this bank. Our conclusion is therefore that the
Crédit Mobilier does not seem to have been supervising them properly
(hence the bankruptcy of the Compagnie Immobilière). The Crédit
Mobilier’s ‘seal of approval’ was probably not worth as much as that of
J.P.Morgan & Co or the Großbanken. De Long in his paper mentioned that
‘J.P.Morgan and Co’s approval of an issue had become…a large factor
which inspires confidence in the investor and leads him to purchase.’ In our
case, the active participation of the bank in a company did not ‘add value’
to the share prices.
Secondly, we evaluated the monitoring role of the bank as regards its
investment policy. This study of corporate finance in France concentrates on
132
GENERAL CONCLUSIONS
133
THE ROLE OF BANKS IN MONITORING FIRMS
unclear. When the bank was founded, Courtois fils, a specialist on financial
questions and author of a Traité élémentaire des operations de bourse et de
change3 wrote:
This definition does not make clear the role and the limitation of such
establishments. The merchant banks are distinct from other banks in the
sense that they are, at least in France, the only institutions to offer the
companies some help in their creation and organisation. The French
merchant bank must evaluate the worth of the potential project that the
firm wishes to undertake. It founds and reorganises the company around
this idea. Its aim is not to keep control permanently: once the enterprise is
well established in its industrial sector, the merchant bank will publicise
its assets, mostly by its introduction on the stock market. The merchant
bank does not expect any further profit from its participations: it will
maintain a very small holding in the company and a representation in the
shareholders’ assembly. This was not how the Crédit Mobilier saw its role:
it was not a merchant bank but an investment bank: participation for the
Péreire brothers signified a source of profit to be exploited in both the
short and long term.
The most serious restrictions to our conclusions are:
The Péreire brothers, with their initiative and their ambitions were
themselves the cause of the liquidity difficulties. The search for immediate
profit led them to make huge investments (e.g. the Compagnie Immobilière).
Henri Germain of the Crédit Lyonnais was more careful:
The Crédit Lyonnais was a long-term work and twenty years of trial
and error were needed by its founder to make its foundations solid.
More lucky than the Péreires, it had the chance and the ability to last. 5
134
GENERAL CONCLUSIONS
135
APPENDIX A
Data on the Crédit Mobilier
The following data are presented in order to understand better the place of
the bank in the financial market of the time. The balance sheet of the Crédit
Mobilier is presented in order to discuss the policy regarding investment and
dividends in the operations of the Crédit Mobilier (cf. Table A1).
The investment account and the cash/balance account show us the power the
bank had over this first period: these two accounts reflect the risky attitude of
the founders. This is reinforced by the dividend distribution (described by the
two last columns and the data below). The dividends declared were considerably
higher than those distributed by other merchant banks during the same period.
Further information on the bank is also disclose by a careful examination
of the share prices fluctuations over the period. Let us first consider the
prices reported in Table A3. These figures reflect the apparent power of the
institution from 1852 to 1867.
Some comments can be made on these prices. In 1856 and 1857 the
fluctuations of the share prices are large; at that time the bank issued
240,000 new shares at a price of 500 francs. In 1862 the share price went
from 705 francs in January to 1,285 francs in October. The general increase
in liquidity in France was probably one reason why lots of speculators
bought shares in financial markets in this period. In 1864 the increase of
liquidity of the Banque de France caused the same phenomenon.
136
Table A1 Balance sheet of the Crédit Mobilier from 1853 to 1866 (in millions of francs)
Source: (1903:111).
138
Table A4 Balance sheet of the Crédit Mobilier from 1903 to 1914 (in millions of francs)
The sources of data on prices are Chadeau (1988:216) and Levy Leboyer
(1970:79–119). The general price index A is composed of the two following
series: B agricultural price index and C industrial index. The repartition
between the two sectors is implied by the weight attached to agriculture and
industry in the GNP. For a detailed presentation see Levy Leboyer’s article
pp. 115–17.
140
APPENDIX B
141
APPENDIX C
Data relative to the affiliated companies
The aim of this appendix is to present the data underlying the analysis
presented in Chapter 2. The share prices between 1852 and 1867 presented
here constitute an insight of the whole data set used to run the regression
and will be provided in order to give an idea of the impact of the Crédit
Mobilier’s bankruptcy on the share prices of companies.
As regards Table A6 no fluctuations in the share prices can be noted for
the set of affliated companies.
142
APPENDIX C
The reorganisation of the Crédit Mobilier after the fall of the Péreires
obliged the society to liquidate part of the claims it had with its partners,
that is the Crédit Foncier.
The revenues of the company during the period were as shown in Table
A7. The expenses were as shown in Table A8.
In March 1881, the Crédit Mobilier reached an agreement with the
liquidators of the Compagnie Immobilière to rebuy some property not yet
liquidated. From this period on, all the claims with the enterprise were
resolved.
Table A9 shows the share price and the dividend distribution between
1855 and 1866.
143
Table A6 Data available for affiliated companies: share prices (francs)
Source: Annales de la Statistique de Paris: Numéro Récapitulatif (1954) pp. 180 et seq. for the railways; private account 65AQ
Archives Nationales de Paris for the insurance company l’Union, the Compagnie Transatlantique, the Compagnie Immobilière, the
Compagnie Parisienne d’Eclairage.
Note: na stands for not available.
APPENDIX C
145
APPENDIX D
Data relative to non-affiliated companies
146
Table A10 Data for non-affiliated companies
148
Table A11 Share prices for affiliated companies for 1866 (francs)
In this appendix, the raw data and its transformation in order to construct
the test will be described. To run the regressions, three parameters must be
available for all companies in the sample: investment, cash flow, and Q. To
estimate this last parameter the book values taken from the balance sheets
have been used. Before reporting any data, some explanations and
clarifications might be made. The investment variable is clearly specified in
the balance sheets and exhibits few changes as regards the actual accounting
.
If we now consider the existing literature on Tobin’s Q, it appears that
most authors take into consideration Q which is the ratio of the valuation of
corporate physical capital in the stock markets to its estimated cost of
reproduction. In the statistical work presented here we do not use this
definition of Q but rather common equity Q. Because of insufficient data on
the market value of debt, we could only approximate ‘Tobin’s Q’ with what
we should be calling strictly speaking ‘common equity Q’. As regards the
definition of the variable K it corresponds to the ‘capital brut’ (gross capital)
which over the period did not include a systematic depreciation rate. In
order to avoid any confusion it will be called Cb.
Table A14, A15 and A16 describe the values of shares that we call p,
investment, gross values of assets Cb and cash flows respectively for the
railway companies, the mining industry and the steel firms. The sample
cover the years 1860–75. As can be seen, some values are missing for
certain years. For the unavailable data, missing values have been substituted.
The situation of the Compagnie Immobilière must be considered
separately. As this company went bankrupt at the same time as the Crédit
Mobilier, little information was available over the period (from 1861 until
1869). However, the introduction of this element in the sample is relevant as
it produces many elements of explanation and interpretation on the bank’s
policy toward investment (the analysis of the results proves that the Crédit
Mobilier did not behave like a Großbanken before the bankruptcy, but
changed policy after 1868). The sources of our information in this special
case are the balance sheets collected in the Archives Nationales de Paris and
152
Table A14 Data available for the railway companies
also some financial newspapers over the period. The data are reported in
Table A17.
All figures represent real values in millions of francs.
Some general comments can be made on Table A17. In comparison to the
first three, we remark that the fluctuations of the parameters are larger and
seem to be difficult to justify. They reflect in some sense the ‘irrationality’
of the Péreires as regards their investment policy. Up to the end (1868
represents a significant date), the bank continued to believe in the
profitability of the company and provided a large amount of liquidity (e.g.
the magnitude of investment i). The test presented for this company alone
will support this argument.
The last variable, Q, is simply the ratio of the market value to book
value. This procedure is virtually identical to the one de Long used in his
work.
As the raw data do not conform to traditional accounting rules, some
transformation has been made. First a depreciation rate δ is introduced for
the book value. The first value noted Cbk1861 is chosen as I1861/δ where δ=5
per cent. Some words of explanation are necessary to justify the choice of
departure for the calculation. If we refer to Jean Bouvier et al. (1965:165–
91) we can note that there was a lack of information for the preceding
period. According to the data available for this work we then choose to
begin in 1860.
The series is then evaluated by:
Cbkt+1 = Cbkt(1 – δ) + It
where d is a constant and chosen such that the series is slightly increasing.
This transformation of the data leads to an increase of Cb.
156
Table A18 New values of Q when introducing a depreciation rate of 5 per cent
THE ROLE OF BANKS IN MONITORING FIRMS
Table A18 exhibits the new value of Q. Some remarks can be made on
these figures. If the series exhibits numbers that are increasing slightly over
the period, the figures obtained for Q are quite different from those obtained
previously. In order to analyse this difference, we propose to see what
would have been the data had another origin been chosen.
Let us consider that the first value is Cbk1862=I1862/δ. The following value
are then obtained by
As the set of data as a whole has the same characteristic (slightly increasing) over
the period we concentrate our argument on the first values for a few companies
(three railway companies, for example). The figures are reported in Table A19.
It appears that the first values have an influence on the data: if we
compare the third and the fourth columns, we see that the figures for Q
are increasing. The same phenomenon is to be noted for the other cases.
However the biggest change comes from the introduction of a
depreciation rate.
The second step of our argument will be to choose other values for d. In
order to see if our choice of d is valuable or not, we evaluate two other
series by considering small variations for d. The idea is then to discuss the
modification of data when choosing a depreciation rate smaller then 5 per
cent and to see the fluctuations of the same parameter when d is larger. The
figures obtained for Q remain low, but fit the theory (i.e. they are smaller
and drecreasing when d decreases.
The last step is to divide all variables by the general price index. Levy Leboyer
(1970:79–119) provides us with a series. As all the figures are expressed in basis
1880, we transpose them in real terms by applying the formula:
158
Table A20 Data in real terms
THE ROLE OF BANKS IN MONITORING FIRMS
160
NOTES
INTRODUCTION
161
THE ROLE OF BANKS IN MONITORING FIRMS
1 Translated from the original: Le sort du Crédit Mobilier n’est pas seulement
caractéristique des rivalités entre groupes financiers et des dangers de la
spéculation boursière: il rend sensible le problèms permanent de l’engagement
industriel.
2 Cf. Baldy (1922). The author provides several examples, among which are the
Crédit Lyonnais and the Banque de l’Union Parisienne. Some explanations are
necessary on what could be understood by ‘control’. The literature on
asymmetric information and incomplete contract presented in the first section
indicates that debt does not provide control. Instead it provides access to
privileged information, which the bank is then able to use in the process of
monitoring and enforcing the contract. Control, however, stays with the
managers of firms. As the Crédit Mobilier provided both debt and equity to its
customers, it can be said that this bank clearly exerted control over the firms
with which it was affiliated.
3 Huth (1918:11).
4 The quotation is taken from a memoir addressed to a group of bankers whose
support the Péreires hoped to obtain.
5 The whole of Bouvier, Furet and Gillet’s Le mouvement du Profit en France au
19ème Siècle provides general information about the fluctuations of share prices
over the period. It gives an indication of the increasing importance of the
financial market over the period.
6 By ‘short-term’ we mean three to six months.
7 See Bouvier, Furet and Gillet (1965).
8 Translation from the original: ‘Notre société a toujours considéré comme un
principe de haute moralité commerciale de ne jamais ouvrir une souscription, de
ne jamais recommander une entreprise sans s’y interesser elle même dans une
forte proportion, et ses administrateurs se sont fait un devoir de s’y associer.’
9 In the balance sheets where these advances were disclosed, the Administrators of
the Crédit Mobilier never mentioned that these advances to companies were not
based on any collateral. ‘Shares and bonds on collateral’, to cover additional
loans, had to be taken into consideration independently of the advances to
companies.
10 Cf. Leland and Pyle (1977).
11 See de Long and Becht (1992).
12 Statutes of the Société de Crédit Mobilier, 8 November 1871. The new bank was
capitalised at 80 million francs with 160,000 shares of 500 francs each; holders
of the old stock obtained two new shares for five old and could subscribe at par
for 48,000 additional shares. The founders of the new bank (including only two
members from the original bank, Charles Mallet and Frederic Grieninger) took
the remainder.
13 Numbers of events happened over the period, including the Spanish-American
war and the US annexation of Puerto Rico. In 1900 the Crédit Mobilier had
162
NOTES
The lender of last resort’s responsibility is to the entire financial system and
not to specific institutions…The lender of last resort exists not to prevent the
occurrence but rather to neutralise the impact of financial shocks…[Its] duty is
a twofold one consisting first, of lending without stint during actual panics and
second, of acknowledging beforehand its duty to lend freely in all future
panics…[It] should be willing to advance indiscriminately to any and all sound
borrowing on all sound assets no matter what the type…In no case should the
central bank accommodate unsound borrowers. The lender’s duty lay in
preventing panics from spreading to the sound institutions, and not in rescuing
unsound ones…. All accommodations would occur at a penalty rate, i.e., the
central bank should rely on price rather than non-price mechanisms to ration
use of its last resort lending facility…The overriding objective of the lender of
last resort was to prevent panic-induced declines in the money stock.
(Humphrey 1975:9)
163
THE ROLE OF BANKS IN MONITORING FIRMS
1 See Chapter 1 with reference to credit rationing literature, and Stiglitz (1981).
2 The debate as regards the advantage of a strong relationship between banks and
industry is one of the oldest in financial economics. The originality of Hoshi,
Kashyap and Scharfstein (1989, 1990, 1991) was to use an econometric
methodology in order to support the argument. Hoshi, Kashyap and Scharfstein
(1989) is a basic reading which constitutes the foundation of the statistical work
presented in this paper.
3 De Long (1989) and Ramirez (1992).
4 Hoshi, Kashyap and Scharfstein (1989, 1991).
5 Annual report: April 1860.
1 The original:
2 These are not the only alleged control mechanisms in a capitalist economy. For
instance, there are evolutionary arguments which suggest that firms which
manage their resources efficiently will survive, while those that do not, will not.
For some criticisms of this argument, see, for instance, Diamond and Stiglitz
(1974), Stiglitz (1982).
3 In this article Stiglitz suggests that the firm should be considered as a problem
of multiple agents and multiple principals.
4 In French, ‘Société anonyme à responsabilité limitée’.
5 Translation from the original:
164
NOTES
L’argent aujourd’hui déposé notre caisse était en grande partie enfoui dans les
tiroirs, sans interêt pour ceux qui le gardaient, sans utilité pour la société.
Mais elles (les sociétés financières) jouent généralement un rôle plus actif:
elles ‘prennent ferme’ un certain ‘paquet’ de rentes ou d’obliga-tions,
‘préparent’ le marché ou attendent quelque circonstance favor-able et font
‘infiltrer’ les titres dans le public; elles ‘transforment’ elles-mêmes les
enterprises privées en sociétés anonymes dont elles émettent ensuite les
actions. Elles vont jusqu’à ‘créer’ des industries ou ‘organi-ser’ un commerce,
utilisant une idée qui leur fut insufflée. Ce sont done les courtiers, les
commissionnaires, les négociants de cette marchandise moderne que l’on
appelle valeurs mobilières: elles facilitent et même organisent le contact entre
l’épargne et ceux qui en ont besoin.
5 GENERAL CONCLUSIONS
165
BIBLIOGRAPHY
166
BIBLIOGRAPHY
Bernanke, B. and M.Gertler (1989) ‘Agency costs, net worth and business
fluctuations’. American Economic Review, LXXIX, 1: 14–31.
Bernheim, D. and M.Whinston (1985) ‘Common marketing agency as a device for
facilitating collusion’, Rand Journal of Economics, 16, 2: 269–281.
—— (1986) ‘Common agency’, Econometrica, 54, 4: 923–943.
Besanko, D. and A.V.Thakor (1987) ‘Collateral and rationing: sorting equilibria and
competitive markets’, International Economic Review 28, 3: 671–689.
Bigo, H. (1947) Les Banques Françaises au cours du XIXème Siècle, Paris.
Blee, E. (1882) ‘La Vérité sur la Crise Financière, Paris.
Bonin, H. (1988a) Histoire Economique de la France depuis 1880, Masson.
—— (1988b) ‘La Grande Depression Française de la Fin du Dix Neuvième Siècle:
Reflexion sur sa datation et sur sa fonction’, Histoire, Economie et Société 6, 4:
509–533.
Bonnet, V. (1865) La Revue des deux Mondes, Paris.
Born, K.E. (1983) International Banking in the 19th and 20th century, Leamington
Spa: Berg Publishers.
Bouvier, J. (1960) ‘Les Péreire et l’affaire de la Banque de Savoie 1860–64’, Cahiers
d’Histoire 3: 383–410.
—— (1968) Naissance d’une banque: le Credit Lyonnais, Paris: Flammarion.
—— (1972) ‘Systèmes bancaires et entreprises industrielles dans la croissance
Européenne du XIXe siècle’, Annales ESC 1: 46–70.
—— (1988) Histoire économique et histoire sociale, Geneva: Librairie Droz.
Bouvier J., Furet F. and Gillet M. (1965) Le Mouvement du Profit en France au
19ème Siècle, Industrie et Artisanat, Mouton and Co.
Brief, R. (1981) ‘Nineteenth century accounting error’, in J.R.Edwards (ed.)
Company Legislation and Changing Patterns Disclosure in British Company
Accounts 1850–1914, London.
Brown, R. (1905) A History of Accounting and Accountants, Edinburgh: T.C. &
E.C.Jack.
Calomiris C.W. and Gorton G. (1991) ‘The origins of banking panics: models, facts
and bank regulation’, in G.Hubbard (ed.) Financial Markets and Financial Crises,
University of Chicago Press / National Bureau of Economic Research.
Cameron, R.E. (1953) ‘The Crédit Mobilier and the economic development of
Europe’, Journal of Political Economy, December, 4: 461–488.
—— (1967) Banking in the Early Stage of Industrialization: A Study in Comparative
Economic History, New York: Oxford University Press.
Cameron, R.E. and V.I.Bovykin (1991) International Banking 1870–1914, New York:
Oxford University Press.
Caron, F. (1973) Histoire de l’Exploitation d’un Grand Réseau de Chemin de Fer: La
Compagnie du Chemin de Fer du Nord 1846–1937, Paris: Mouton.
Carosso, V.P. and Sylla, R. (1991) ‘US banks in international finance’, in R.Cameron
and V.I.Bovykin (eds) International Banking, 1870–1914, New York and Oxford:
Oxford University Press, 48–71.
Cassis, Y. (1992) Finance et Financiers in European History 1880–1960, Cambridge
University Press.
Chadeau, E. (1988) ‘L’Economie Nationale aux XIXème et XXème siècles’,
Annuaire Statistique de l’Economie Française aux XIXème et XXème siècles,
Presses de l’Ecole Normale Supérieure, (1, 3).
167
THE ROLE OF BANKS IN MONITORING FIRMS
168
BIBLIOGRAPHY
169
THE ROLE OF BANKS IN MONITORING FIRMS
Innes, R.D. (1990) ‘Limited liability and Incentive contracting with ex ante action
choices’, Journal of Economic Theory, 52: 45–67.
Jacquillat, B. and M.Levasseur (1984) ‘Signaux, mandats et gestions financière: une
synthèse de la littérature’, Finance, 5: 31–63.
Jeidels, D. (1905) ‘Das Verhältnis der deutschen Großbanken zur Industrie mit
besondere Berücksichtung der Eisenindustrie’, in Staats- und
Sozialwissenschaftliche Forschungen, Band XXIV, Heft II: , 1–271, Leipzig:
Duncker & Humblot.
Jensen, M.C. and W.C.Meckling (1976) ‘The theory of the firm: managerial behavior,
agency costs and ownership structure’, Journal of Financial Economics, 3: 305–
360.
Johnston H.T. and R.S.Kaplan (1987) Relevance Lost: the Rise and Fall of
Management Accounting, Boston, Massachusetts: Harvard Business School Press.
Jorgenson, D. (1971) ‘Econometric studies of behavior: a review’, Journal of
Economic Literature, IX: 1111–47.
Kaufmann P. (1914) La Banque en France, Paris.
Keely, M.C. (1990) ‘Deposit insurance, risk about market power in banking’,
American Economic Review, 4: 1183–1200.
Kindleberger, C.P. (1984) ‘Banking and industry between the two wars: an
international comparison’, Journal of European Economic History, 39, 2: 7–26.
Kocka, J. (1978) ‘Entrepreneurs and Managers in German Industrialization’, in
Mathias P. and M.Postan (eds) The Cambridge Economic History of Europe, Vol.
VIIThe Industrial Economics: Capital Labour and Enterprise Part I, Cambridge:
Cambridge University Press.
Kose, J. and A.Kalay (1982) ‘Costly contracting and optimal payout constraints’,
Journal of Finance, XXXVII, 2: 373–390.
Krause, D. (1991) ‘Zur Geschichle der Mitteldeutschen Creditbank’, Bankhislorisches
Archiv, 17: 17–41.
Kreps, D. and R.Wilson (1982) ‘Reputation and imperfect information’, Journal of
Economic Theory, 27: 253–279.
Kreps, D., P.Milgrom, J.Roberts and R.Wilson (1982) ‘Rational cooperation in the
finitely-repeated prisoner’s dilemna’, Journal of Economic Theory, 27: 245– 252.
Laffont, J.J. (1987) ‘Le risque moral dans la relation de mandat’, Revue Economique,
41, 1: 5–23.
Laffont, J.J. and X.Freixas (1988) ‘Contrats Bancaires Optimaux’, Toulouse:
GREMAQ.
Laffont, J.J. and E.Maskin (1986) ‘The theory of incentives: an overview’, in
W.Hildenbrand (ed.) Advances in Economic Theory, Cambridge University Press.
Lambert, R. (1983) ‘Long term contracts and moral hazard’, Bell Journal of
Economics, 14, 2: 441–452.
Landes, D.S. (1956) ‘Vieille banque et banque nouvelle: la révolution financière du
XIXème siècle’, Revue d’Histoire Moderne et Contemporaine, 3: 204–220.
Leland, H.E. and D.H.Pyle (1977) ‘Informational asymmetries, financial structure and
financial intermediation’. Journal of Finance, 32, 2: 381–387.
Lenoir, M. (1919) ‘Le mouvement des cours des valeurs mobilières françaises depuis
1856’, Bulletin de la Statistique Générale de la France, Octobre, 65–92.
—— (1922) ‘Les valeurs mobilières et la crise de 1920 (indices économiques
compares en France et divers pays)’, Bulletin de la Statistique Générale de la
170
BIBLIOGRAPHY
171
THE ROLE OF BANKS IN MONITORING FIRMS
172
BIBLIOGRAPHY
Shiller, R. (1981) ‘Do stock prices move too much to be justified by subsequent
changes in dividends?’, American Economic Review, 71, 2: 421–36.
—— (1989) Market Volatility, Cambridge, MIT.
Smirlock, M., Gilligan T. and W.Marshall (1984) ‘Tobin’s q and the structure
performance relationship’, American Economic Review, 74, 4: 1051–1060.
Solow, R.M. (1982) ‘On the lender of last resort’, in C.P.Kindleberger and J.P.
Laffarge (eds) Financial Crises: Theory, History and Policy.’ Cambridge
University Press.
Stiglitz, J.E. (1985) ‘Credit markets and the control of capital’, Journal of Money,
Credit and Banking, 17: 133–152.
—— (1982) ‘The inefficiency of the stock market equilibrium’, Review of Economic
Studies, 49(2), April, 241–261.
Stiglitz, J.E. and A.Weiss (1981) ‘Credit rationing in markets with imperfect
information’, American Economic Review, 71: 393–410.
Stole, L.A. (1990) ‘Mechanism design under common agency’, Mimeo, MIT,
November (unpublished).
Teichova, A. (1988) ‘Rivals and partners: banking and industry in Europe in the first
decades of the twentieth century’, Uppsala Working Papers in Economic History,
Report from the Vienna Banking Industry Symposium.
Thornton, H. (1802) An Enquiry into the Nature and Effects of the Paper Credit of
Great Britain, edited by F.A.Hayek, Fairfield Augustus: M.Kelley.
Tilly, R. (1966) Financial Institutions and Industrialization in the Rhineland 1815–
70, London: University of Wisconsin Press.
—— (1986a) ‘German banking 1850–1914: development assistance for the strong’,
Journal of European Economic History, 15, 1: 113–152.
—— (1986b) ‘Germany 1815–1870’, in R.Cameron, et al. (eds) Banking in Early
Stages of Industrialisation: a study in comparative economic history New York,
London, Toronto: Oxford University Press.
Tobin, J. and W.R.Brainard (1977) ‘Asset markets and the cost of the capital’, in
Balassa B. and Nelson R. (eds) Economic Progress, Private Values and Public
Policy: Essays in Honor of William Fellner, North Holland.
Webb, D. (1987) ‘Conflict of interest, bond rationale and the value maximisation
rule’, Economica, 54, 3: 455–463.
Weber, A. (1903) Die rheinisch-westfälischen Provinzialbanken und die Krisis,
Leipzig.
Weston, J.F. and T.E.Coppeland (1987) Financial theory and corporate policy, 3rd
edn, Addison Wesley Publishing Company.
Whale, B.P. (1930) Joint Stock Banking in Germany: a study of the German Credit
Banks before and after the War, London: Macmillan and Co.
Wilson, C. (1977) ‘A model of insurance markets with incomplete information’,
Journal of Economic Theory, 16: 167–207.
Wrück, K.H. (1990) ‘Financial distress, reorganization and organizational efficiency’,
Journal of Financial Economics, 27, 2: 419–444
Yoshikawa, H. (1980) ‘On the q theory of investment’, American Economic Review,
3: 739–743.
173
INDEX
174
INDEX
175
THE ROLE OF BANKS IN MONITORING FIRMS
176
INDEX
177
THE ROLE OF BANKS IN MONITORING FIRMS
Ramirez, C. 79, 86, 93, 95, 110, stock market and banks 51–2
116–17 Stole, L.A. 13
Riesser, J. 78 supervisory role of banks 15–24, 109–31;
risk-neutrality in common agency 14 association with, advantages and
Ross, S. 12 disadvantages 116–24; bank finance
and market finance 19–20; banking
Saint Simon, Henri, comte de 119 panic in bankruptcy 114–16; and
Sayous, A.E. 122, 125, 126 cartels 23; and contracts 21–3; and
Schaaffhausen’schen Bankverein 27 Crédit Lyonnais, comparisons 120–3;
Scharfstein, D. 77, 79, 80, 115, 133 credit policy in 110; finance 16–19;
Schartz, A. 112 financial crisis in bankruptcy 111–14;
Schneider (company) 44, 67; France and Germany, comparisons
bankruptcy, impact of 68–70; data on 124–8; intermediation and
147 commitment 23–4; and investment
Seillière, Baron 29 performance of firms 23; model of
share prices: affiliated companies 60–4; 17–18
and bankruptcy of Crédit Mobilier Sylla, R. 51–2
112; fluctuations 34; indices, France
140–1; monthly, affiliated companies Terrenoire (company) 44, 67;
148–51; non-affiliated companies bankruptcy, impact of 68–70; data on
67–71 147
shareholders on bankruptcy of Crédit Thalmann and Co. 37
Mobilier 128–9 Thornton, H. 38–9
Shiller, R. 45, 47 Tilly, R. 22, 125
signalling theory 1; basic concepts Tobin, J. 82, 91
8–12; model of 11
Société de l’Hôtel et des Immeubles de Union Fire 55–9; data on 143
la Rue de Rivoli 31, 32 l’Union (insurance company) 43;
Société d’Eclairage Électrique 43 bankruptcy of Crédit Mobilier,
Société des Machines Outils 44 impact of 55–9; data on 143
Société des Omnibus de Paris 43 United States: monitoring role of banks
Société Immobilière 34 51; stock markets, volatility tests on
Société Metallurgique de Montbard- 49–50
Aulnaye 43
Société Parisienne d’Eclairage 41 volatility ratios 49–52
Société Parisienne d’Eclairage et de
Chauffage par le Gaz 43
Société Raffineries Say 43 Wallut, C. 36
Solow, R.M. 39 Whale, B.P. 127
Stiglitz, J.E. 112 Wrück, K.H. 118
178