The Impact Of Lending Relationship On Risk Premiums Finance Essay
This paper presents empirical evidence from a representative sample of Tunisian firms on the importance of loan officers in the production of ‘soft information’ by opposition of ‘hard’ information, testing whether firm borrowers have a relationship with their bank or with their loan officer. To establish the main determinants of the credit conditions in Tunisia, our study is based on a sample of 297 companies, dealing with a Tunisian bank from 1998 to 2002. The main finding of this paper is that the risk premium depends only on the hard information proxies. However, the availability of credit seems related to the interaction between the loan officer and the customer through the collection of ‘soft’ information.
Biographical notes: Meryem BELLOUMA is an Associate Professor of Finance at the University of Economics and Management (Nabeul) Tunisia. Her research interests are banking relationship, credit decision, corporate governance and corporate finance. Her research has been published in journals such as Revue Internatioanle PME, Journal of Emerging Market Review, and Revue des Sciences de Gestion. Member of of the research laboratory “Finance et Stratégie des Affaires (FIESTA)”.
Sami BENACEUR is an Associate Professor of Financial Economics at the High Institute of Commerce (Carthage) Tunisia. His research interests are corporate finance, Economics reforms and Financial Macroeconomics. His research has been published in journals such as Applied Economic letters, International Review of Finance, Economic Notes, Journal of Comparative Economics, Journal of Economics and Business, International Review of Financial
Analysis, Research in International Business and Finance, Frontiers in Economics and Finance, Managerial Finance…Founder member of the laboratory “Economie et Finance Appliquées”.
Abdelwahed OMRI is a Professor of finance at the High Institute of Management (Tunis) Tunisia. His research interests are corporate governance, financial information, Finance Market and banking. His research has been published in journals such as Revue Française de Gestion, Revue Internatioanle PME, Journal of Emerging Market Review, and Revue des Sciences de Gestion. President of the research laboratory “Finance et Stratégie des Affaires (FIESTA)”.
The Tunisian banking system has experienced several changes, which have profoundly modified their structure. The abolition of functional separation of institutions, the shift from structural to prudential regulation and the ongoing revolution of information technology have changed substantially the Tunisian banking environment, enhancing competition and putting banks relationship under pressure. Today, the Tunisian banking industry comprises 14 commercial banks, 6 investment banks, 3 merchant banks and 8 off-shore banks and it is organized western style  .
Besides, the Tunisian firms are vulnerable because of their dependency on financial institutions for external funding (Bellouma et al., 2005 b). These firms simply do not have access to public capital markets. As a result, they try to develop a lending relationship with their bank. Under lending relationship, banks acquire information over time through contact with the firm, its owner, and its local community on a variety of dimensions and use this information in their decisions about the availability and terms of credit to the firm (e.g., Boot, 2000, Grunert et al., 2005 and Berger and Udell, 2002, Ginès and Pedro, 2007).
The purpose of this study is to identify the impact of ‘soft’ and ‘hard’ information generated by the lending relationship on the risk premium and the credit availability from a Tunisian dataset. We try to clarify whether the process of bank lending in Tunisia, is based on ‘hard’ information, or depends on the loan officer (‘soft’ information). In fact, it is generally left unspecified whether the primary relationship is between the bank and the firm or between the loan officer and the firm’s owner, from which the bank acquires and stores the relationship information, and how this information may be disseminated within the bank.
As far as we are concerned, this study is the first paper that focuses on the estimation of credit availability and risk premium in a developing country using the two types of information. As one observe, all the existing papers use data from the US and European economies from which lessons are not directly applicable to an emerging market economy like Tunisia. In fact, the banks in Tunisia are characterised by their dependency on main bank which determinates the level of credit allocation and control the risk premium. Besides, in Tunisia the capital market is not developed and firms are particularly financed by credit. The econometric approach is also interesting which is based on the context of panel data analysis. Finally, it is done on a period where the Tunisian banks try to manage relationship lending in order to overcome negative effect.
The rest of the paper is organized as follows. Section 2 briefly provides the literature review. Section 3 describes the research design and details the methodology adopted. Section 4 exposes the findings. Finally, section 5 concludes.
II. Literature Review
Bank relationship is beneficial for developing economies, where markets and firms are smaller and legal protection is rather weak as well as there is little transparency. In fact, given underdeveloped capital market, bank relationship should be stimulated because of necessity to intensify efforts towards a successful restructuring process (Bellouma and Omri, 2008).
The relationship occurs when banks get involved in firm activities in order to obtain customer-specific information, often proprietary in nature. Based on the acquired information banks are able to evaluate efficiency of their funds allocation (Bellouma et al., 2005 a).
Relationship banking aims mainly at resolving information asymmetries and agency problems which are very severe in transition economies. The more severe information asymmetries and contract incompleteness, the higher the advantages of relationship banking should be (Bellouma et Omri, 2004 and Berger et Udell, 2006).
More precisely, the lending relationship is associated with the collection of two types of information. The ‘soft’ information may not be easily observed and verified by others, or transmitted to others. In fact, this information is difficult to quantify for independent analysis and relies on the proximity between the loan officer and the customer (Scott, 2003). As a consequence, relying on this type of information necessitates various organizational checks. In contrast, the ‘hard’ information is based on relatively objective criteria, such as financial ratios and statements. This type of information can be reduced to a series of numbers. A second dimension of ‘hard’ information is the way in which it is collected. The collection method does not need to be personal. Instead the information can be entered into a form without the assistance or significant guidance from a human data collector.
Recent empirical evidence provides support for the importance of lending 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). These studies use duration of the banking relationship, the scale of the relationship (e.g., the number of lenders), and the scope of the relationship (e.g., the number of products used as the primary quantitative proxies for the strength of the relationship). For example, Berger and Udell (1995); Degryse and Van Cayseele (2000); Harhoff and Korting (1998), Petersen and Rajan (1997); and Cole (1998) have frequently found that both credit availability and loan terms improve as the duration of banking relationship increases which can be considered as an implicit measure of the proximity between the loan officer and the customer. Angilini and al. (1998) did not find any association between the length of relationship and loan rate paid for a sample of Italian firms, nor did Elsas and Krahnen (1998) and Harhoff and Korting (1998) for a sample of German firms. This finding can be interpreted as an evidence of the influence of ‘hard’ information on the lending decision.
However, empirical work that documents the effect of improved ‘soft’ information on credit availability, loan terms, or the role of the loan officer in producing ‘soft’ information is limited. Harhoff and Korting (1998) use an indicator variable called ‘mutual trust’ from their survey of German small firms and find a negative relationship between trust and cost of borrowing and incidence of collateral, but no significant relationship with credit availability (although signed positively).
Scott (2003) proposes a more direct test of the importance of the proximity of the loan officer as the focal point for the collection of ‘soft’ information about opaque credits. Two variables are used that can serve as a proxy for the role the loan officer plays in the accumulation of ‘soft’ information. The first is the loan officer turnover, which directly addresses the role of loan officers in the accumulation of ‘soft’ information, where increased turnover should result in destruction of ‘soft’ information. The second is the social contact rating, which can be thought of as one channel for loan officers to obtain ‘soft’ information. The author find that increased loan officer turnover is hypothesized to destroy ‘soft’ information; while a lower social contact rating should be associated with less production of ‘soft’ information.
Data have been obtained from the credit portfolio of a Tunisian bank, for which firm total asset is greater than 500 million and less than 1 billion dinars over the period 1998-2002. We have only excluded financial institutions which have special characteristics notably different credit activities. It includes 297 unlisted and independent companies observed from 1998 to 2002, on which financial data are collected.
There is no hope to investigate seriously the impact of ‘hard’ and ‘soft’ information on credit availability and loan rates granted by the bank if we limit ourselves to information coming from the annual financial statements. For that purpose we have designed a questionnaire and submitted it to all loan officers, in order to take into account the dimension of the ‘soft’ information.
Before presenting the variables’ measurement, we try to reveal some characteristics of the sample used. The panel is mainly composed of limited liability companies (on average, during the five periods of the study, they present 79.4% of the companies included in the sample). The limited corporations represent only 20.6%. Moreover, more than 3/4 of the companies in our sample operate in the service sector.
Besides, we note that the loan officers study occasionally the customer’s demand of credit jointly. This feature provides support for the idea that the benefits of a relationship lie with the loan officer who produces the ‘soft’ information, not the bank.
The hard aspects of information generated by the lending relationship were measured using the firm characteristics (size and age), the contract terms (the type of guarantees) and the financial variables (the leverage and the quick ratio). The two control variables retained are the juridical status of the surveyed companies and their sector. We retain also the duration of the banking relationship, the existence of a saving account and the uniqueness of the customer relationship.
In order to capture the social aspect of the bank lending relationship, the questionnaire includes the social contact which can be thought as the transmission mechanism of ‘soft’ information from the small firm owner to the loan officer (Scott, 2003). The mutual trust is the second ‘soft' information proxy. It shows the borrower’s willingness to provide information and to inform the loan officer about problems. Mutual trust can be regarded as a mechanism approximating the impression of the loan officer as for the stability of the relation and informational flows (Harhoff and Körting, 1998 and Lehmann and Neuberger, 2002). Besides, we include the turnover of the loan officer, the mean of communication. All the variables included in the estimations are summarized in Table 1.
--------- Insert Table 1 -------
In order to identify the main determinants of the risk premium and the availability of credit, we expose in this sub-section the several assumption isolated from the theory.
First specification tries to determine the influence of the ‘hard’ information variables (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC, SERV) isolated from the theory of financial intermediation on the risk premium.
Hypothesis 1: The loan rate decreases with the firm’s size, age and the level of guarantees provided and increases with the leverage and liquidity constraints.
The second specification combines the ‘soft’ and ‘hard’ factors resulting from customer relationship. More precisely, we estimate the effect of (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC, SERV, SOCIL, MTRT, OREL, SVAC, EXCL, WEKT, COMMf and DWEK) on the risk premium.
Hypothesis 2: The risk premium is weak for firms with an exclusive, a longer duration and intensive bank relationship.
The third specification tries to determine the influence of the ‘hard’ information variables (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC, SERV) isolated from the theory of financial intermediation on the availability of the credit.
Hypothesis 3: The loan availability increases with the firm’s size, age and the level of guarantees provided and decreases with the leverage and liquidity constraints.
The fourth specification combines the ‘soft’ and ‘hard’ factors (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC, SERV, SOCIL, MTRT, OREL, SVAC, EXCL, WEKT, COMMf and DWEK) resulting from customer relationship.
Hypothesis 4: The loan availability is high for firms with an exclusive, a longer duration and intensive bank relationship.
As mentioned earlier, the main objective of this paper is to identify the determinants of the risk premium and the availability of credit in Tunisia. We are estimating panel regressions, assuming a one factor random effects specification. This enables us to use the full sample while controlling for unobserved heterogeneity among individuals  . Since the availability of credit is a censored variable we use a Tobit-formulation for our third and forth regressions like Lehmann & Neuberger (2002).
Specifically, we try to estimate the following models:
PREMit= a1 + b1 HARD it +c1 CONTOL it + eit (1)
Where HARD is composed by (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC) and CONTROL (CORP, SERV)
PREMit= a2 + b2 HARD it + c2 SOFT it+c2 CONTOL it + eit (2)
Where HARD is composed by (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC) and SOFT is composed by (SOCIL, MTRT, OREL, SVAC, EXCL, WEKT, COMMf and DWEK) and CONTROL (CORP, SERV).
AVLBit= a3 + b3 HARD it+c3 CONTOL it + eit (3)
Where HARD is composed by (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC) and CONTROL (CORP, SERV)
AVLBit= a4 + b4 HARD it + c4 SOFT it +c4 CONTOL it + eit (4)
Where HARD is composed by (SIZE, AGE, LEVR, ARINV, MNGR, PLDG, QKRT, LLIC) and SOFT is composed by (SERV, SOCIL, MTRT, OREL, SVAC, EXCL, WEKT, COMMf and DWEK) and CONTROL (CORP, SERV)
Before we turn to the test of the different hypotheses, we present some descriptive statistics concerning variables used in our analysis. The characteristics of their distributions directly stand out: The average distribution of the risk premium is 2.81% which is higher than the one used by Berger and Udell (1995). We note through the measure of the credit availability that on average, 4.26% of the commercial effects are rejected. Compared with Harhoff and Körting (1998), who report an average leverage rate of 37.9%., Tunisian’s leverage rate in our sample seems to be lower at 21.55%. The quick ratio represents on average the value of 1.21. This indicator reflects the capacity of refunding to the company, in short term. The value of this financial indicator, as exposed by Berger and Udell (1995), reaches, however, 2.51.
--------- Insert Table 2 -------
The first set of columns of table 3 shows that the risk premium is a fraction of the financial leverage. This variable is intended to identify the risk of the firm. As expected, firms having the biggest debts pay proportionally a higher risk premium. The explanation that we can advance is that banks may consider a levered firm as a risky one and pay more attention to this parameter which is a source of ‘hard’ information. Besides, the type of guarantees is also an explanatory element of the risk premium charged by banks. A firm with a higher level of collateral (MNGR) will obtain a cheaper credit than a firm with a lower level of guarantees. The intuition behind this finding is that less risky firms will then choose a contract with lower interest rate but increase the guarantee level. On the contrary, more risky firms will not be willing to choose such a contract, knowing that their risk level is higher and, therefore, that the probability of guarantee execution is higher.
The coefficient attached to the variable (SIZE) proves to be significant at the level of 1% for the three specifications selected. Thus, we notice that the size factor constitutes a dominating factor of risk differentiation between firms in our sample. Besides, the positive correlation between size and the risk premium means, that at equal risk, the bank is able to increase its margin, whenever the size of the company is high. This result is in line with Repetto et al. (2002), which implies that banks have the possibility to extract additional rents, in order to cover the possible losses recorded with small and medium-sized firms. Indeed, the imperfection of the market involves, in period of recession, an extreme prudence of the banks as the level the companies’ opacity tends to be accentuated. Thus, the bank grants a low interest rate for the potentially risky companies (of low size), in order to avoid any possible deviation of their behaviour. Therefore, the increase of the risk premium is likely to encourage the borrowers to undertake riskier projects.
The last variable that explains the risk premium charged by the bank is the quick ratio. It is worth noting, however, that this ratio is positively linked to the level of risk premium. This finding seems to support the explanation we have advanced concerning the impact of the SIZE. Thus, an improvement of the solvency of companies seems to induce an increase of the interest rate charged by banks in order to benefit from the better shape of the firm .
Finally, the regression presents empirical evidence on whether the loan officers have an important role in the production of soft information. The different proxies used for the intensity of soft information production by loan officers are found to have no significant effect. Thus we reject the assumption that the benefits of a relationship lie with the loan officer who produces soft information, not the bank. In other words, for a given level of business risk, soft information accumulation cannot reduce management risk. That’s why the bank is not permitted to reduce risk ratings and thus cannot improve the risk premium.
--------- Insert Table 3 -------
We now estimate the effect of hard and soft information on the availability of credit. As the table 4 shows (first column), the size of the firm, its financial leverage, the quick ratio and the offer of several guarantees remains the core determinants of the availability of credit.
Given the choice of the measure of the dependant variable (AVLB), we can specify whether the firm is rationed by the bank. In fact, the negative correlation between the quick ratio (QKRT) and the credit availability shows that the bank rations firms that were able to meet their engagement. In other words, this can be evidence towards the seniority of bank debt. In this case, firms might need new funding but asset substitution problems might be rampant, particularly in financial distress states. Thus, the bank can extend resources to the firm in order to overcome the financial deterioration.
This result is also supported by the interaction between the leverage ratio and the availability of credit. In fact, the negative sign shows that the bank tends to give financial support to the firm that are more levered. Therefore, we can stipulate that this kind of relation reveals the “dark side” of relationship banking (its costs) or is consistent with the soft-budget constraint problem. This friction has to do with the potential lack of toughness on the bank’s part in enforcing credit contracts that may come with relationship-banking proximity. The key question is whether a bank can credibly deny additional credit when problems arise. That is, a borrower on the verge of defaulting may approach the bank for more credit to forestall default. While a de novo lender would not lend to this borrower, a bank that has already loaned money may well decide to extend further credit in the hope of recovering its previous loan.
Concerning the AGE of the firm, which can reflect a measure of the firm’s investment opportunities, the regression shows that, typically younger firms are more rationed than the older. In fact, the age of the firm can be considered like a measure of the degree of the opacity of the firm. Then the firms that do not have sufficient public component of information find themselves more rationed by the bank.
Before turning to the second specification, we note that the SIZE of the firm and the modality of the guarantees (ARINV) are positively linked to credit availability. One explanation can be advanced in this respect, is that the bank compensates the registered loss on its relation with smaller firms, which are not able to provide the bank with sufficient collateral, by rationing the larger firms. Thus, banks compel the larger firm, and eventually are able to provide the requested guarantees, to meet their engagement and to pay back the financial and trade credit at time.
The second column of table 4 reports in addition to the hard information proxies, the influence of the closeness of a firm to its bank. The mutual trust emerging between the loan officer and the firm seems to have a significant impact on the availability of credit. In fact, this interaction variable plays a key role in the development of the relationship. This factor may emanate from the positive experience between the loan officer and the firm’s manager in the past. It will lower the moral hazard and hence monitoring cost. It proves, as shown in the table 4, to have a significant negative impact on rationing. This finding can be explained by the fact that the nature of the relationship between the two parties influences the quantity and the quality of information available to the bank. In close relationships the loan officer understands the operating environment of a particular business. He receives signals concerning managerial attributes and firm’s prospects. Thus, the relationship provides the basis of understanding customer needs.
The mutual trust rating can be thought of as one channel for loan officers to obtain soft information, i.e., the social interaction between the firm’s owner and the loan officer. A higher mutual trust rating should, be associated with more production of soft information. In other word, the increased trust rating should result in a lower chance of rejection or unfulfilled credit needs.
Loan officers play a critical role in acquiring soft information about the management risk of the firm, information that cannot easily be transmitted within a banking organization, unlike historical default experiences offered by many commercial sources or the accumulation of financial statements over time. In addition, soft information production appears to have a more persistent effect on the ability of firms to get flexible financing. Overall, the empirical results support the idea that banking relationships reside with the loan officer.
--------- Insert Table 4 -------
The purpose of this article is to assess the role of the soft and hard information in the Tunisian context where the problems of the informational asymmetry impede decisional process by the bank.
A lending relationship depends on transactions as well as interactions between the bargaining partners. The empirical analysis of bank lending has confirmed that risk premium depends only on hard information elements. This result confirms the characteristics of the Tunisian bank system which rely on the main bank who fixes the vitality of loan rate. Then, the role of the loan officer seems to be neutral in adjusting the terms of the loan. However, our analysis confirms that the availability of credit is influenced not only by firm characteristics, but also by social interactions between the loan officer and the manager. As a result, policies makers should control the allocation of banks funds in order to avoid hazard problem and behaviour deviation of the loan officer and the firm.
Being close to one’s customers is likely to facilitate a loan officer’s collection of soft information. Because the loan officer is the repository of this soft information, agency problems are created throughout the bank organization. In other words, relationship lending requires that more authority be given to the loan officer, who has the greatest access to the soft relationship information. Besides, relationship lending differs from transaction-based lending technologies that are based on “hard” information that may more easily be observed, verified, and transmitted. This greater authority, in turn, creates agency problems within the bank that necessitate various organizational checks and balances. We offer a modest first step toward addressing this gap in the literature by examining relationship lending within the context of a simple model of the lending function.
Also, given that Tunisian financial system is a bank-based system (the financial market don’t play a key role), the nature and the role of social interaction should be captured by more detailed variables. Besides, it could be interesting to analyse the value of soft information production to small firms and lenders.
Need help with your literature review?
Our qualified researchers are here to help. Click on the button below to find out more:
In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: