The relationship between a commercial bank and its customers is not standardized. Contrary to the other transactions, the signing of a credit contract does not mean the end of the relationship between the two parties. Rather, it marks the beginning of their commitment.
However, several circumstances might emerge once funds have been released that may alter the borrower's ability to pay back the loan.
Such occurrences are increasingly present in the Tunisian context where the information available to the commercial bank and the customer is often asymmetric. This can be easily noticed especially among companies that do not keep adequate records or those that invest in new markets.
This informational disparity urges on the part of the Tunisian bank, a rethinking of the customer's relationship, which requires the development of an adequate communicational structure. Besides, the need to reduce negative effects of this imbalance becomes inevitable because of the evolution of the customers' expectation, on the one hand, and the growing competition among the Tunisian banking industry, on the other. In this regard, the bank risk management relies on the informational identification that is assigned to the customer, which in turn is the prerogative of an organization that collects, produces and manages information (Rivaud, 1995).
This survey is motivated by the diverging debates on the role of the customer's relationship in a context of informational asymmetry. It seeks to enlarge the scope of previous empirical studies that try to understand the banker's behaviour as it approves the credit.
In particular, we study customer's relationship of Tunisian firms'. We attempt to shed light on the different factors that explain the degree of importance of the customer's relationship, established between the bank and the Tunisian companies. We examine whether the characteristics of the firms retained and the conditions at which a given firm obtains loans influence the customer's relationship. According to theory, the consequences of those factors are not clear. These outcomes, in turn, have distinct implications for both market performance and policy. The empirical assessment of these effects is therefore especially valuable. The issues we examine in this paper are important in their own right for the functioning of the financial market, particularly regarding credit access of small and medium size firms in an emerging market.
Recent empirical evidence provides support for the importance of customer relationship to small businesses in terms of both credit availability and credit terms such as loan interest rates (e.g., Petersen and Rajan, 1994, 1995, Berger and Udell, 1995, Cole, 1998, Elsas and Krahnen, 1998 and Harhoff and Körting, 1998). However, empirical work that documents the determinants of the customer relationship is limited. Besides, as one observe, all these papers use data from the US and European economies from which lessons are not directly applicable to an emerging market economy like Tunisia. In addition to that, there is a few studies done in an emerging market (Reppetto et al., 2002).
The rest of the paper is organised as follows. First, it focuses on the role of the customer's relationship in the informational acquisition. Then, taking into accounts the economies of scale, it will address the monopolistic power that the bank can exert to the detriment of the customer (section II). It will attempt to identify the customer relationships' determinants in the Tunisian context (section III). The conclusion is presented in section IV.
II. THE ROLE OF THE CUSTOMER'S RELATIONSHIP IN THE MITIGATION OF THE INFORMATIONAL ASYMMETRY PROBLEM
The banking literature shows that if the borrowers have more information than the lenders, the credit market deteriorates. In fact, in such case, no interest rate can allow the banks to make a profit. Especially, the increase in the interest rate tends to eliminate the solvent companies and to promote investment in riskier projects (Stiglitz and Weiss, 1981).
This situation occurs when bank's access to information is limited and when information gathering is expensive, especially during the earlier transactions. By contrast, the cost of collecting information can decrease when the transactions are repeated continually. Actually, each time the bank renews customer's credit contract; it acquires more information about the firm's past and its state of solvency. This allows the bank to reduce its control and hence the cost of follow-ups. Besides, in this situation the costs of credit contracts negotiations will decrease.
Petersen and Rajan (1994) point out the crucial role of the services offered by the bank in addition to the credit approval, in the information collection that can serve as a durable input for the decision making process as concerns the concession of risk lines over several periods of time. Hence the economies of scale on the informational production (Battacharya and Thakor, 1993 and Thakor, 1995) provide the bank with a means to overcome informational asymmetry problem (Lelands and Pyles, 1977; Campbells and Kracaws, 1980; Fama, 1985 and Diamond, 1991). Then, the company that maintains a narrow (i.e. well established) relationship can benefit from advantageous credit contracts conditions (e.g. low interest rate...) (Boot, 2000, Berger and Udell, 2002).
Petersen and Rajan (1994) point out the possibility created by the customer's relationship in the reinforcing of subsequent controls exerted by the bank, once the customer engages in opportunistic actions. Indeed, the bank reacts to such behaviour by threatening to deprive customer of future financings or other facilities. Such a scenario is possible if the bank has a market power that competition could not neutralize its effect, and that allows it to reshape appropriately the terms of its contracts. In a competitive market, the bank finds itself compelled to accept the relaxation of controls on the customer, especially as the customer can take advantage of better conditions offered by a rival bank.
The market power decreases the information propagation, which make it possible to reinforce the bank's capacity to anticipate the opportunistic behaviours of its clientele and sharpens negotiation skills (Keeley, 1990).
According to Von Thadden (1995), once the credit contract is signed, the two parties find themselves trapped in a situation of bilateral monopoly, providing them with simultaneous gains. In fact, the bank collects some temporary early information, because the companies can put an end to this relation. On its part, the firm could, of course, benefit from a set of services and advantageous credit conditions, under the constraint that a negative signal would be issued if it changes banks.
On the whole, the narrow and continuous interaction with the company assures for the bank a low risk credit. In other words, it improves the quality of the information collected on the one hand, and spreads out the fixed costs ensuing from its production over an important number of services, on the other (Lehmann and Neuberger, 2002).
The effect of the customer's relationship on the interest rate remains ambiguous. Its determination depends on the level of competition among banks and also the way that the bank acts to obtain the requisite information.
Thus, if the information is available and can easily be checked, then the inter-banking competition is likely to limit the margin of every bank. By contrast, if it is not easily controlled, the bank whose relation with the customer is intense (continuous in time), acquires a monopolistic power that no other potential bank can have (Scott, 2003, Elsas, 2005).
Geenbaum, Kanatas and Venezia (1989), Sharpe (1990) and Rajan (1992) support the statement that the customer's relationship has less impact to reduce interest rate than the changes in the bank's cost of providing the credit.
Fisher (1990) and Sharpe (1990) found no significant impact of the customer's relationship on the interest rate. In fact, the borrowers' bankruptcy entails losses for the bank. To compensate for these losses, the bank imposes restrictions on the solvent customers through an increase of the credit interest rate. The bank's negotiating power with the customers grows with the monopolistic dominance.
This behaviour on the part of the bank can be explained by the fact that within a competitive market, it can neither expect to continually share the company's future surplus, nor extend its customer's relationship. Therefore, the uncertainty of the firm's perspectives is likely to increase the interest rate charged to cover potential losses resulting from failure to pay back the loans.
Thus, the customer's relationship will have no meaning because it is constructed on the principle that the bank invests in its customer's account, while expecting that the firm pays it back subsequently by continuing transactions with the bank. However, the more competitive is the market; the lower is the chance to recover the initial investment. This implies that the bank's likelihood to develop such a relation in a similar context is reduced since the credit market competition can stop the profits gained from its establishment (Hoshi, Kashyap and Scharfstein, 1990 Petersen and Rajan, 1994). This result supports that the proliferation of competition in the banking market is very often accompanied by a decline in the value of the customer's relationship.
In the following section, we attempt to shed light on the different factors that explain the degree of importance of the customer's relationship, established between the bank and the Tunisian companies.
III. THE EXPLANATORY POWER OF THE DETERMINANTS OF LENDING RELATIONSHIPS
We present, in this section, the sample of the selected companies, the variables utilized in our model and the results obtained in our study.
1. Data and Sample Constitution
The information used in the survey comes from the entire credit relations maintained by a Tunisian commercial bank from 1998 to 2000. This sample contained 76 unlisted companies.
One segment of clientele in the sample is composed of industrial and commercial private companies operating in different economic sectors. The priority in terms of credit within this segment should be given to firms which present a good risk level, maintain a high volume of foreign tade, and are engaged in diverse activities. The other segment includes private firms presenting a good risk level, having at least three years of activity, issuing at least three balance sheets, having achieved an important foreign trade volume, having reached a minimum annual revenue of 5 million TNDS (Tunisian Dinars, 1TND= 1,4 US- $), and having satisfactory collaterals.
We have excluded from the sample all public industrial or commercial firms operating in different sectors of the economy. These companies have special characteristics notably a relatively low risk which results in a reduced margin and a narrow financial visibility. Also, we have excluded banking and non banking institutions whose potential credit activities remain limited.
2. The variables
In order to identify the determinants of the lending relationship, we assume that the informational asymmetry is due to the operating risk (RISK). In addition to this variable, we assume that risk plays an important role in defining customer relation. Additionally there are some control variables that also have a role and we will introduce them shortly.
The control variables such as the interest rate (INTE), the collateral (COLL), etc, are introduced in the model to capture their effects on the dependent variable customer's relationship (CUST).
CUST : The customer's relationship (CUST) will be measured by the ratio of mid as well as long term bank debts over total bank debts. CUST is our dependent variable. Generally, a bank which proves a certain trust vis-à-vis that trusts a particular customer can easily accept a long-term relation and can avoid frequent short-term credit negotiations. Values of CUST that are close to 1 indicate long-term relation.
RISK : The variable risk is measured by the difference between the minimum of the cash flow of last year and the cash flow of this year divided by the total company's equity during the first retained year. This ratio indicates the insolvency risk and the company's inability to pay back its debts. We expect that the risk assumed by the bank, after having agreed to a credit, decreases as the customer's relationship reinforces itself. In fact, the bank acquires more information on the risk profile of its client as the relation is maintained in time.
TYPE : The firm's legal status can be measured by two binary variables one for corporation (TYPE1) and one for limited liability (TYPE 2) . The first Dummy variable is equal to 1 if the business is a corporation, and 0 if otherwise. The second Dummy variable is equal to 1 if the business is a limited liability, and 0 if otherwise.
This factor reduces the shareholders' and executives' responsibility and imposes, at the same time, specific constraints on the company's equities. Intuitively, the customer's relationship is expected to be more intense within the limited liability company, because of the narrow interaction between the company's manager and the bank.
.
LEVE : The leverage is measured by the ratio of borrowed funds over the company's equity. Therefore, we expect that a high level of indebtedness (large value of LEVE) is associated with a higher risk of failure, which in turn can affect the customer's relationship default.
SPEC : This variable measures the asset's specificity. We define it as the ratio of intangible assets over the total assets.
The intangible assets such as good will creates some problems for credit granting institutions, since the intangible assets may not be redeployed, hence altering the bank's ability to liquidate them in case of default. This conjuncture lets us assuming that companies with specific assets should maintain an intense customer's relationship in order to benefit from the bank's credits.
MATU : This variable measures debt maturity. It is defined as the ratio of the current liability over long term liability. A small value of MATU corresponds to a long maturity and can be interpreted as a sign of the company's performance as it reflects company's past liabilities and its reputation. So we expect that this variable (given the measure retained) is negatively related to the dependant variable.
SECT : The sector can be measured by two binary variables one for the industrial sector (SECT1 )and one for the services sector (SECT2). The first Dummy variable is equal to 1 if the company belongs to the industrial sector, and 0 if otherwise. The second Dummy variable is equal to 1 if the company belongs to the services sector, and 0 if otherwise. These sectors are well known by the bank and can influence its decision concerning credit granting.
COLL : This variable indicates the type of collateral used in support of the loan. The borrower's background constitutes a useful source of information to which the bank can resort to evaluate the necessary guarantee. When the client does provide substantial information through its relationship, he expects to receive more favourable treatment regarding the collaterals.
The first Dummy variable is equal to 1 if the loan is without collateral (COLL1), and 0 if otherwise. The second Dummy variable is equal to 1 if the loan is secured by a mortgage (COLL2), and 0 if otherwise. The third Dummy variable is equal to 1 if the loan is secured by a pledge (COLL3), and 0 if otherwise. The fourth Dummy variable is equal to 1 if the loan is secured by interdependent guarantees (COLL4), and 0 if otherwise. The fifth Dummy variable is equal to 1 if the loan is secured by several collateral (COLL5), and 0 if otherwise.
INTE : This is a measure of interest rate. It is defined as the ratio of the total cost of borrowing over the total of company's bank debts. The level of interest rate must be consistent with the company's specificity and the bank's lending policy. In fact, a high interest rate tends to eliminate the solvent borrowers (the problem of adverse selection) or to encourage the companies to invest in riskier projects (the problem of moral hazard) ( Stiglitz and Weiss, 1981). Thus, we expect that by maintaining an intense customer's relationship with its bank, the company is likely to experience a decrease in the interest rate.
SIZE : The firm size is measured by the logarithm of the total assets. The activities of the large-size companies are often varied and their success does not depend on one single project. Unlike the modest-size companies, the risk that larger companies present is not so significant
3. The Model Specification and Test Results
To identify the variables that explain variations in lending relation, we have used a regression model utilizing the entire data set. More precisely, we regress, using estimation technique on panel data with fixed effect, customer's relationship [CUST] on the variable measuring the potential risk [RISK] and all the control variables [CONT] (legal status, leverage, assets' specificity, debt maturity, activity, collateral, interest rate and size). Hence, the model is defined as follows:
CUSTit= a + b RISK it,t+k + Σcj CONTit j+e it (1)
Where, i and t indexes correspond respectively to the company and to the time period. The index j refers to various control variables as listed above. The index (t, t+k) indicates that the economic risk may be revealed at the time t or at a later time t+k. The index j is associated to the control variables included in the analysis. Table 1 shows some descriptive statistics of the variables used in our analysis.
Table 1 : Descriptive Statistics of the Variables
Variables
N
Mean
St.Deviation
Skewness
Kurtosis
Significance
CUST
76
0.314
0.288
0.353
-1.152
0.000***
RISK
76
0.347
4.443
13.035
168.817
0.000***
INTE
76
0.402
0.988
6.412
48.194
0.000***
MATU
76
22.586
101.782
8.609
84.613
0.000***
SPEC
76
0.388
0.816
18.155
381.978
0.000***
LEVE
76
1.127
29.857
10.512
131.727
0.000***
[SECT1]
76
0.1859
0.3892
1.6118
0.618
0.000***
[SECT2]
76
0.7178
0.4503
0.1411
0.3483
0.000***
[TYPE=0: limited liability company]
76
5.039E-02
0.2189
1.618
2.271
0.000***
[TYPE =1: corporation]
76
0.1288
0.3351
-0.970
-1.992
0.000***
[COLL=5: Several Collateral]
76
0.1411
0.3483
2.065
4.086
0.000***
[COLL=4: Interdependent Guaranty]
76
0.1120
0.3155
4.118
14.988
0.000***
[COLL=3 : Pledge]
0.5274
0.4995
2.220
1.6987
0.000***
[COLL=2: Mortgage: 2]
76
0.3589
0.1399
-0110
-1.596
0.000***
[COLL=1: without Collateral]
76
0.6795
0.2165
1.517
2.376
0.000***
*** Significance <1%.
The above table indicates that the spread of the debt maturity around its average is wider than those of the other variables. This observation implies that the standard deviation of this variable is higher. This is due to the fact that in our sample we had some extreme positions where some companies had only short-term debts and few others had only long-term debt. As to the leverage, although its standard deviation is smaller than that of the debt maturity, it is still large enough to signify that the level of debts varies from one company to another.
Table 2 : Results of the regression (1)
Dependant variable: CUST
Variables
Coefficients
Standard
Deviation
Signification
Constant
0.621
0.219
0.005
[SECT=1: industrial sector]
5.622E-02
0.054
0.297
[SECT=2: services]
-
-
-
[TYPE=0: limited liability company]
6.707E-02
0.035
0.054
[TYPE =1: corporation]
-
-
-
[COLL=5: Several Collateral]
-1.749E-02
0.034
0.609
[COLL=4: Interdependent Guaranty]
7.337E-02
0.041
0.076
[COLL=3 : Pledge]
-1.236E-02
0.048
0.795
[COLL=2 : Mortgage: 2]
0.149
0.058
0.010
[COLL=1: without Collateral]
-
-
-
SIZE
-1.764E-02
0.012
0.156
RISK
1.534E-02
0.007
0.035
INTE
-5.183E-02
.015
0.001
MATU
-1.197E-04
0.000
0.350
SPEC
0.101
0.046
0.027
LEVE
-2.006E-03
0.001
0.067
As the table 2 shows, the sector, the size, the maturity and the collateral under the modality of 'Several Collateral', 'Pledge' aren't significant. Besides, the collateral under the modality 'Interdependent Guaranty' and the leverage are significant at the level of 10%.
Regarding the variable RISK, a p-value of 0.035, the alternative hypothesis is accepted, which means the risk coefficient b is significantly different than zero. Moreover, it is positively correlated with the customer's relationship. We can explain this by the fact that risky Tunisian companies try to establish an intense customers' relationship allowing them to access bank credit.
This result contradicts the one presented by Petersen and Rajan (1994), which states that less risky businesses develop narrow relations with their banks.
Taking this into consideration, it is interesting to determine the nature of the compensation expected by the bank. In other words, does the bank consider an increase in the interest rate or rather a stronger guarantee, or may be both? To respond to this question, we analyse the impact of the interest rate and the different forms of guarantees on the customer's relationship. More precisely, we ran a second regression, in the second model we left out some of those variables that were retained in the first model.
CUSTit = a + b INTE it, + Σcj COLLitj +e it (2)
Where, i and t indexes correspond respectively to the company and to the time period. The index j refers to various control variables as listed earlier. The index j is associated to the control variables included in the analysis.
Table 3 : Results of the regression (2)
Dependant variable: CUST
Variables
Coefficients
Standard Deviation
Signification
Constant
0.696
0.171
0.000
INTE
-5.576E-02
0.013
0.000
[COLL=5: Several Collateral]
4.062E-02
0.054
0.011
[COLL=4: Interdependent Guaranty]
-3.282E-02
0.063
0.600
[COLL=3 : Pledge]
-9.550E-02
0.062
0.125
[COLL=2 : Mortgage]
0.112
0.146
0.444
[COLL=1: without Collateral]
-8.361E-04
0.070
0.990
The interest rate (INTE) charged by the bank proves to be a significant variable influencing the customer's relationship, as shown by the table 3. The answer to the previous question is not affirmative because the interest rate and the customer's relationship are negatively correlated. Therefore, the bank does not compensate for the increase in the risk following the intensification of the customer's relationship by a rise in the interest rate.
This observation agrees with the one stated by Petersen and Rajan (1994) who pointed out the ties between the interest rate and the customer's relationship (measured by the length of relation, number of banks partners etc.…). The conclusion reached following their study is based on the fact that the interest rate is simultaneously affected by the customer's relationship. It's also closely related to the findings of Rajan (1992) and Thakor (1996) which showed that an intense relation between the customer and the bank generally implies the availability of the useful information to take decisions, and consequently a reducing the effect of the informational disparity. It also captures reputation effects by the borrower.
Besides, the intensity of this relation generates no effect on the rise of credit cost. Thus, we reject the notion of the 'hold up problem' resulting from the bank monopolistic power (a corollary of the establishment of the customer's relationship (Sharpe, 1990)). This idea allows raising a crucial dimension of the customer's relationship resulting from its durability, because the more the company reimburses its debts, the more it reinforces its viability and out of this experience, the bank will be able to reduce the interest charges.
By establishing such a relation, the Tunisian companies that apply for a credit, hope for a lower interest rate. Indeed, when the firm faces a prohibitive rate, it may try to discipline the bank by resorting to non-bank credit that might even be more expensive. The bank is then penalized by the fact that the demand by good firms turns up to be lower than the one expected in a situation of monopoly.
This result is logical since it specifies the reaction of a firm facing the monopolistic power exerted by the bank to its detriment. Indeed, it seems unrealistic that the bank expropriates a company without an immediate reaction on the part of the customer.
By examining the coefficient of the interest rate, it appears that the customer's relationship is influenced by the charged interest rate, but whose magnitude of effect is weak; which seems to be consistent with the following two theoretical explanations.
The first argument is that the customer's relationship is an important criterion, but that the companies prefer more important credit availability rather than a decrease in the interest rate. The second explanation is that the bank is not pressured by the size of the market and by the intensity of the competition which constrains it to reduce the interest rates on the granted credits. This is due to the benefit of a monopolistic power reinforced by the relation that the bank maintains with its clientele.
The results obtained reject the first type of compensation, consisting of the increase of the interest rate following the intensification of the risk assumed by the bank and induced by the amplification of the relation. Taking this into account, we check the second type implying an increase in collateral requirement.
Among the five types of guarantees only number 5 (COLL =5; several collateral) has a coefficient that is statistically significant with a p-value of 0.011.
The positive correlation between COLL5 and CUST suggest that as collateral increases so does the customer relation.
However, other values of COLL prove to be far from being significant ('Interdependent Guaranty', 'Pledge', 'Mortgage' and 'without Collateral').
This is despite the fact that the coefficients of 'Pledge' and 'Interdependent Guaranty' have the correct sign, which is a negative correlation with the customer's relationship.
We conclude that the compensation obtained by the bank due to the increase in the risk level, would not be an increase in the interest rate, rather it would be a requirement of several guarantees.
This fact presume that the bank included in the study, attempted to avoid the sanction that it could face following the increase of the interest rate, especially as the current Tunisian situation is characterized by the increase in the competition among the rival banks.
After having assessed the impact of the interest rate, the risk and the guarantee on the customer relationship, the impact of the other variables of control is considered [Table 2].
The obtained results show that assets specificity (SPEC) has a direct influence on the dependent variable. This means that the more specific are the assets controlled by the company, the more the bank requires the establishment of a strong customer's relationship to provide a credit. As for the economic sector, it is not a variable that explains the relation. This result confirms the recent statistics obtained from the credits received by business sectors. Indeed, the credits volume granted to the services sector presents 49.8% against 40.2% to the industrial sector.
It appears that the legal status (TYPE) under the limited liability company mode, explains the customer's relationship. This type of company tends to reinforce its customer's relationship via the narrow and continuous interactions with the bankers.
In the same way, the leverage (LEVE) seems to explain the variation of the customer's relationship. The higher the leverage, the lower is the chance of any credit agreement.
The size of the company does not constitute an explanatory variable for the customer's relationship. In the Tunisian environment, all companies, irrespective of their size, look to establish such a relation in order to take advantage of the bank credits, since banks are their main external source of financing.
Furthermore, the informational asymmetry between these companies and the market creates an uncertainty regarding expertise, solvency and the nature of the investment opportunities that they present. Therefore, as the Tunisian businesses rarely borrow on the open market, a bigger part of their funds come from the banks.
Finally, the maturity of the debts (MATU) is not an explanatory variable for the customer relationship. It can be misleading to follow the argument that the agreement on a short and long-term credit requires the establishment of such a relation.
IV. CONCLUSION
The purpose of this article is to assess the role of the customer's relationship in the Tunisian context where the problems of the informational asymmetry impede decisional process by the bank. Two main conjunctures, related to the importance of the customer's relationship, have been developed by the literature on banking
The first has impact on the nature of information generated through a continuous and lasting interaction over time. Indeed, the need for a strong guarantee and the setting an adequate interest rate are not the only means that allow the creditors to overcome the problem of informational asymmetry. The company's background and its reputation also constitute a useful source of information which the bank can take into account when evaluating credit applications.
Thus, the relationship banking has a distinct role to play and can be a value-enhancing intermediation activity.
The second conjuncture is related to the impact of the economy of scale of information production on the increase of the monopolistic power of the bank.
This problem, known as 'Hold up problem', is likely to lead to a prohibitive increase in the interest rate charged by the bank, especially in a situation where a market power can be exercised easily.
Indeed, as the bank requires some information on its customer, it will be able to use it to client's detriment, because when the company changes banks, it will leave a negative impression on their image.
At the conclusion of the empirical study, we found that the bank meets the problems of the informational asymmetry regarding credit granting, by collecting information in a gradual manner through the development of customer's relationship.
More importantly, we demonstrated that an intense customer's relationship is rewarded by an offer of an interesting interest rate for the borrower. But, it is accompanied by an increase in the level of guarantee requirement.
Such behaviour can among others, explain the positive correlation between the customer's relationship and the risk. Indeed, the compensation sought by the bank because of the increase of the risk is to be covered by stronger guarantees, to reduce the moral hazard likely to emerge in an ex-post perspective.
At this level, it is important to see whether the customer's relationship affects the availability of funds. In other words, could the company, establishing a close relation with its bank, decreases the amount of credits that bank grants?
The response to this question requires future investigation alleviating the limits of this study by identifying whether the informational advantage acquired through the maintenance of an intense customer's relationship is used by the bank to alter the amount of guarantee rather than the level of interest rate.
Another limitation, relates to the choice of the retained variables. Indeed, the results are obtained from the accounting data that are more or less contaminated with errors based on the information appearing on the credit contracts.
Such limitations should result in research on data and data gathering in a more strict manner and initiate work on contracted data, in order to contribute to a better understanding of the banks' strategy in an environment characterised by lack of reliable information. Such developments help to appreciate the risk assumed by the bank
However, the granting of a credit is fundamentally a balancing act between the present and the future: today the bank issues the credit amount (a certain amount) and hopes that the customer will be able to pay it back (uncertain future). The risk is about a prognosis of the future of the company.
REFERENCES
Battacharya, S. and Thakor A.V., (1993), "Contemporary Banking Theory", Journal of Financial Intermediation, Vol. 3, (2) pp.2-50.
Berger A., and Udell G.F., (1995), "Relationship Lending and Limes of Credit in Small Firms", Finance Journal of Business, Vol. 68, (1) pp. 351-381.
Berger, A., and Udell, G., F. (2002). « Small Business Credit Availability and Relationship Lending: The Importance of Bank Organisational Structure», Economic Journal, Vol 95 (2), pp.319-325.
Boot, A., (2000), "Relationship banking: what do we know?" Journal of Financial. Intermediation Vol 9 (3), pp. 7-25.
Campbell, T. and Kracaw, W., (1980), "Information Production, Market Signalling, and the Theory of Intermediation", Journal of Finance, Vol. 35, (2) pp.863-882.
Cole, R., (1998), "The Importance of Relationships to the Availability of Credit", Journal of Banking and Finance, Vol. 22, (1) pp.959-977.
Diamond, D., (1991), "Monitoring and reputation: The Choice between Bank Loans and Directly Placed Debt", Journal of Political Economy, Vol. 99(4), pp.689-721.
Elsas, R. (2005), "Empirical determinants of relationship lending", Journal of Financial Intermediation, Vol. 14, (1), pp.32-57
Elsas, R. and Krahnen, J.P., (1998), "Is Relationship Lending Special from Credut-Fila Data in Germany", Journal of Banking and Finance, Vol. 22, (2)pp.1283-1316.
Fama, E.F., (1985), "What's Different about Banks?" Journal of Monetary Economics, Vol.15, (1) pp. 29-39.
Fisher, K. (1990), "Lending Relationship as a Link Instrument's between Banks and Businesses: A Theoretical and Empirical Analysis", working Paper of Bonn University, pp. 12-15.
Greenbaum, S., Kanatas, G. and Venezia, I., (1989), "Equilibrium Loan Pricing under the Bank-Client Relationship", Journal of banking and finance, Vol.13 (2), pp. 221-235.
Harhoff, D. and Körting, T., (1998), "Lending Relationships in Germany-Empincal Evidence from Survey Data", Journal of banking and finance, Vol. 22, (1) pp.1317-1353.
Hoshi, T., Kashyap, A. and Scharfstein D., (1990), "Bank Monitoring and Investment: Evidence from the Changing Structure of Japanese Corporate Banking Relationships", Journal of Financial Economics, Vol.25, (1) ,pp.99-122.
Keeley, M. C., (1990), "Deposit Insurance, Risk, and Market Power in Banking", American Economic Review, Vol. 80(5), pp. 1183-1200.
Leland, H. and Pyle D., (1977), "Information Asymmetries, Financial Structure, and Financial Intermediaries, Journal of Finance, Vol. 32, (3) pp.371-387.
Lehmann, E., and Neuberger, D., (2002), « Do Lending Relationship matter? Evidence from Bank Survey Data in Germany», Journal of Economic Behaviour and Organization, Vol. 45, (2), pp. 339-359.
Petersen, M. and Rajan R., (1994), "The Benefits of Lending Relationships: Evidence from Small Business Data," Journal of Finance, Vol. 49, (2), 3-37.
Petersen, M. and Rajan, R., (1995), "The Effect of Credit Market Competition on Lending Relationships", Quarterly Journal of Economics, Vol. 110 (2), pp. 407-443.
Rajan, R. G., (1992), "Insiders and Outsiders: The Choice between Informed and Arm's Length Debt", Journal of Finance, Vol.47, (2), pp.1367-1400.
Repetto, A., Rodriguez, S., and Valadès, R., (2002), « Bank Lending and Relationship Banking: Evidence From Chilean Firms», Working Paper, pp.9-17.
Rivaud D., (1995), "Le rationnement du Crédit et l'Incertitude", Revue d'Economie Politique, (105), Vol (2), , pp.226-247.
Scott, J.A. (2003), « Bank mergers and small firm financing », Journal of Money, Credit and Banking, vol. 35, (2), p. 7-29.
Sharpe, S. A., (1990), "Asymmetric Information, Bank Lending, and Implicit Contracts: A Stylized Model of Customer Relationships," Journal of Finance, Vol. 45, (1), 1069-1087.
Stiglitz, J.E., and Weiss, A., (1981), "Credit Rationing in Markets with Imperfect Information," American Economic Review, Vol. 71, (3)pp.393-410.
Thakor, A.V., (1995), "Fiancial Intermediation and the Market for Credit", Handbooks in Operations Research and Management Science, Vol. 9, Amsterdam, Elsevier, pp. 1073-1103.
Thakor, A.V., (1996), "Capital Requirements, Monetary Policy, and Aggregate Bank Lending: Theory and Empirical Evidence", Journal of Finance, Vol. 51, (3) pp. 279-324.
Von Thadden E. L. (1995), « Long-Term Contracts, Short-Term Investment and Monitoring», Review of Economic Studies, Vol. 62, (1) pp. 557-575.
