Abstract
Social banks are financial intermediaries paying attention to non-economic (i.e., social, ethical, and environmental) criteria. To investigate the behavior of social banks on the credit market, this paper proposes both theory and empirics. Our theoretical model rationalizes the idea that reciprocity can generate better repayment performances. Based on a unique hand-collected dataset released by a French social bank, our empirical results are twofold. First, we show that the bank charges below-market interest rates for social projects. Second, regardless of their creditworthiness, motivated borrowers respond to advantageous credit terms by significantly lowering their probability of default. We interpret this outcome as the first evidence of reciprocity in the credit market.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
Own calculations based on the figures in GABV (2012).
We refer to the figures of the European Federation of Ethical and Alternative Banks (FEBEA) available on www.febea.org.
We henceforth use “social bank” to describe any bank claiming to pay attention to extra-financial criteria, regardless of their specific nature, be they social, ethical, or environmental. Arguably, a triple bottom line may be advocated (Global Report Initiative 2011) insofar as social banks often combine ethical and environmental concerns. Akin to other works on socially responsible lending (e.g., Gutiérrez-Nieto et al. 2011; Allet and Hudon 2003), we consider environmental concern as part of social concerns. Moreover, Norman and MacDonald (2004) state that the triple bottom-line rhetoric may be misleading and act as a smokescreen.
Becchetti et al. (2011) identify the following foundational principles of social banks: (1) awareness of non-economic consequences, (2) access to finance as a human right, (3) efficiency and probity, (4) fair redistribution of profits, (5) full transparency, (6) encouragement of active involvement of shareholders and savers in decision making, and (7) ethical inspiration in all activities.
Cooperative status affects not only the design of the institution's governance but also the capital structure of its balance sheet. Ferri et al. (2010) and Iannotta et al. (2007) show that financial cooperatives tend to be better capitalized than commercial retail banks. Plausibly, this set-up is stronger in social banks. Management can use the diffuse ownership structure to easily retain earnings within the bank (Périlleux et al. 2012). This strategy is in line with the investors’ commitment to forgo financial returns in exchange for the accomplishment of the bank’s social mission. In addition, the cooperative status helps aligning the managers’ behavior with the bank’s social mission (Kitson 1996). Becchetti and Huybrechts (2008) draw the same conclusion for fair trade organizations.
Each ABS shareholder must remain below the 3-percent voting-right threshold. Triodos Bank’s shares are held in trust by an ad-hoc foundation, whose board is appointed by depository receipt holders with limited voting rights.
We only consider the two key categories of stakeholders: investors (shareholders, savers) and borrowers, and disregard other categories such as the staff. Nevertheless, Cornée et al. (2012) show that employees of social banks exhibit higher social preferences than their counterparts working in mainstream banks.
Paradoxically, more evidence is available on microfinance institutions active in developing countries than on social banks active in developed countries. The existing evidence on microcredit activity is, however, not transposable to social banking because the microcredit lending methodology is specific. It is based on the supply of standardized small loans without collateral (Armendariz and Morduch, 2010). Microfinance institutions typically charge identical interest rates to most—if not all—borrowers, and simply tailor loan size to their borrowers’ perceived creditworthiness (Agier and Szafarz 2013a).
Actually, our model includes a homogenous group of opportunistic borrowers (zero cost of cheating) and a continuum of motivated borrowers characterized by their degree of motivation, defined by their cost of cheating.
See www.lanef.com.
In 2010, its deposits-to-assets ratio was 85.92 % and its loans-to-assets ratio was 40.12 %, which is quite low. However, the resources not directly used for loans are entrusted to Le Crédit Coopératif, a partner cooperative bank sharing La Nef’s social values. In 2010, this represented 35.76 % of the balance sheet (La Nef 2010).
The data were collected in November 2008. The sample period for loan granting stretches from January 1, 2001 to November 25, 2004. The November 2004–November 2008 period is used only as a feedback period.
Since September 2007, La Nef has operated four branches.
The Ile-de-France, Provence-Alpes-Côtes-d’Azur and Rhône-Alpes regions are overrepresented since they include the three largest French cities: Paris, Marseille and Lyon, respectively.
Most likely, our sample does not suffer from a selection bias. The missing loans were excluded by accident, not on purpose. Unfortunately, we had no access to information on the denied applications. This in turn limits the possibility of observing the bank's full selection process.
Due to data unavailability, some statistics have been obtained from reduced samples. Location and loan officers are known for 367 firms, age and firm status for 369, and turnover and staff for 55.
The relatively low loans-to-assets ratio (40.12 %) may derive from a scarcity of social projects that break even.
Loan-loss provisioning is governed by law. Therefore, we rule out the possibility that loans with different social ratings are treated differently by the bank.
Admittedly, this argument would be stronger if we had access to data on denied loans, which is unfortunately not the case. Instead, we rely here on the assumption that the loan selection is made within a pool of applications large enough to allow the bank to make unconstrained choices. Although this assumption is debatable, we see no realistic scenario that would make the observed zero correlation spurious.
These percentages are obtained from a sub-sample of 367 firms.
The overall evolutions of the FIN and SR variables are stable. This excludes the possibility for the shift in spreads being driven by a change in the composition of the clientele.
We have also estimated a model explaining the spread. The estimation results are similar to those in Table 5 (“Impact of Social Rating on Interest Rate” section), regarding signs, amplitudes, and levels of significance. However, explaining the spread rather than the interest rate is detrimental to the quality of fit.
The loading of SR in specification (4) is lower than in the previous specifications. Presumably, this is because, unlike FIN ratings, the SR ratings are determined somewhat subjectively by loan officers.
The loans are extended for periods varying from 1 to 20 years. This four-year convention, fixed by the bank, is thus somewhat arbitrary. Still, 87 % of defaults occur within the four years following credit granting.
Logit estimations (not reported) bring similar results.
Sensitivity analysis reveals that variations in this parameter have little effect on the estimates of the NBR t ’s.
Loans in default are non-performing loans at least 90 days in arrears. Actually, LLP can also be manipulated strategically. For instance, banks have incentives to use provisions to manage earnings and regulatory capital as well as to signal information about future prospects (Ahmed et al., 1999). Nevertheless, working with differential—rather than absolute—costs likely offsets any strategic biases.
We use specification (3) rather than specification (4) in order to carry out the analysis on the full sample.
The robustness checks are carried out on the reduced sample for which we have full information (367 firms).
Even though La Nef has several branches, it has a single nationwide credit committee. This committee is composed of two persons: a headquarters-based manager and the loan officer. Importantly, branch-based loan officers take active part in the committee’s decision making. They can communicate all the relevant soft information either by being on-site or by phone. Since the headquarters are located in the South-East branch, loan officers from that branch perhaps influence the credit conditions more than their colleagues from other branches.
In other social banks, the social assessment is carried out according to distinct procedures. For example, in Banca Etica (Italy), a thorough social audit is conducted by the so-called “social auditors or experts”, who are cooperative members trained by the bank.
The provisioning rate of a loan in default is equal to LLP/loan size.
To obtain this figure, we have combined two sources of information. First, Robert de Massy and Lhomme (2008), mention that on average 15.97 % of total staff in French banks are devoted to the screening of SME loan applicants. Second, from annual reports (Banque Populaire de l'Ouest 2010; Crédit Agricole Ille-et-Vilaine 2010; Crédit Mutuel Arkéa 2011) of regional branches of the three major French cooperative banks dealing with SMEs we estimate their numbers of SME loans per officer: 41.09, 35.25, and 33.68 for Banque Populaire de l’Ouest, Crédit Agricole Ille-et-Vilaine, and Crédit Mutuel Arkéa, respectively. Averaging these figures yields 36.67 loans granted per officer per year. This computation is somewhat heroic since the activity sector, the type of clientele, and the lending technology should be held constant.
We used the following instrumental variables: ENVIRONMENT (dummy variable taking value 1 if the borrowing firm works in the environmental sector, and zero otherwise), RURAL (dummy variable taking value 1 if the borrowing firm is located in a rural area, and zero otherwise), NONPROF (dummy variable taking value 1 if the borrowing firm is a not-for-profit organization, and zero otherwise), UNLIMITED (dummy variable taking value 1 if the borrowing firm is an unlimited company, and zero otherwise), CONSORTIUM (dummy variable taking value 1 if the borrowing firm belongs to a consortium, and zero otherwise), and the duplicates (STARTUP and RELATIONSHIP).
References
Adbulkardiroglu, A., & Bagwell, K. (2013). Trust, reciprocity, and favors in cooperative relationships. American Economic Journal: Microeconomics, 5, 213–259.
Agier, I., & Szafarz, A. (2013a). Microfinance and gender: Is there a glass ceiling on loan size? World Development, 42, 165–181.
Agier, I., & Szafarz, A. (2013b). Subjectivity in credit allocation to micro-entrepreneurs: Evidence from Brazil. Small Business Economics, 41, 263–275.
Ahmed, A. S., Takeda, C., & Thomas, S. (1999). Bank loan loss provisions: A reexamination of capital management, earnings management and signaling effects. Journal of Accounting and Economics, 28, 1–25.
Akerlof, G. A., & Kranton, R. E. (2000). Economics and identity. Quarterly Journal of Economics, 115, 715–733.
Akerlof, G. A., & Kranton, R. E. (2005). Identity and the economics of organizations. Journal of Economic Perspective, 19, 9–32.
Allet, M. & Hudon, M. (2013). Green microfinance. Characteristics of microfinance institutions involved in environmental management. CEB Working Paper No. 13/005. Université Libre de Bruxelles.
Armendariz, B., & Morduch, J. (2010). The economics of microfinance (2nd ed.). Cambridge, MA: MIT Press.
Armendariz, B., & Szafarz, A. (2011). On mission drift in microfinance institutions. In B. Armendariz & M. Labie (Eds.), The handbook of microfinance (pp. 341–366). London: World Scientific Publishing.
Banque Populaire de l’Ouest. (2010). Annual report.
Barigozzi, F., & P. Tedeschi. (2011). Credit markets with ethical banks and motivated borrowers. Working Paper DSE No 786. University of Bologna.
Basel Committee on Banking Supervision. (2001). The new basel capital accord. Consultative Document.
Becchetti, L., & Garcia, M. (2011). Do collateral theories work in social banking? Applied Financial Economics, 21, 931–947.
Becchetti, L., Garcia, M., & Trovato, G. (2011). Credit rationing and credit view: Empirical evidence from loan data. Journal of Money, Credit and Banking, 43, 1217–1245.
Becchetti, L., & Huybrechts, B. (2008). The dynamics of fair trade as a mixed-form market. Journal of Business Ethics, 81, 733–750.
Benabou, R., & Tirole, J. (2010). Individual and corporate social responsibility. Economica, 77, 1–19.
Benedikter, R. (2011). Social banking and social finance. New York: Springer.
Berger, A. N., Rosen, R. J., & Udell, G. F. (2007). Does market size structure affect competition: The case of small business lending. Journal of Banking & Finance, 31, 11–33.
Berger, A. N., & Udell, G. F. (1995). Relationship lending and lines of credit in small firm finance. Journal of Business, 68, 351–381.
Boot, A. W. A. (2000). Relationship banking: What do we know? Journal of Financial Intermediation, 9, 7–25.
Brown, M., Fehr, E., & Zehnder, C. (2009). Reputation: A microfoundation of contract enforcement and price rigidity. Economic Journal, 111, 333–353.
Brown, M., & Zehnder, C. (2007). Credit reporting, relationship banking and loan repayment. Journal of Money, Credit and Banking, 39, 1883–1918.
Chakraborty, A., & Hu, C. (2006). Lending relationships in line-of-credit and non-line-of-credit loans: Evidence from collateral use in small business. Journal of Financial Intermediation, 15, 86–107.
Chen, Y., & Li, S. X. (2009). Group identity and social preferences. American Economic Review, 99, 431–457.
Cornée, S., Masclet, D., & Thenet, G. (2012). Credit relationships: Evidence from experiments with real bankers. Journal of Money, Credit and Banking, 44, 957–980.
Cowan, C. D., & Cowan A. M. (2006). A survey based assessment of financial institution use of credit scoring for small business lending. Office of Advocacy, US Small Business Administration.
Cowton, C. J. (2002). Integrity, responsibility and affinity: Three aspects of ethics in banking. Business Ethics: A European Review, 11, 393–400.
Cowton, C., & Thomspon, P. (2000). Do codes make a difference? The case of bank lending and the environment. Journal of Business Ethics, 24, 165–178.
Crédit Agricole Ille-et-Vilaine. (2010). Annual Report.
Crédit Mutuel Arkéa (2011). Annual report.
Defourny, J. (2001). From third sector to social enterprise. In C. Borzaga & J. Defourny (Eds.), The emergence of social enterprise (pp. 1–18). London: Routledge.
Degryse, H., & Van Cayseele, P. (2000). Relationship lending within a bank-based system: Evidence from European small business data. Journal of Financial Intermediation, 9, 90–109.
Dufwenberg, M., & Kirchsteiger, G. (2004). A theory of sequential reciprocity. Games and Economic Behavior, 47, 268–298.
Elsas, R., & Krahnen, J. P. (1998). Is relationship lending special? Evidence from credit-file data in Germany? Journal of Banking & Finance, 22, 1283–1316.
Fehr, E., & Fischbacher, U. (2002). Why social preferences matter—the impact of non- selfish motives on competition, cooperation and incentives. Economic Journal, 112, 1–33.
Fehr, E., Gächter, S., & Kirchsteiger, G. (1997). Reciprocity as a contract enforcement device: Experimental evidence. Econometrica, 65, 833–860.
Fehr, E., & Schmidt, K. (1999). A theory of fairness, competition and cooperation. Quarterly Journal of Economics, 114, 817–868.
Fehr, E., & Schmidt, K. (2003). Theories of fairness and reciprocity—evidence and economic applications. In M. Dewatripont, Hansen, L., & S. Turnovsky (Eds.), Advances in economics and econometrics—8th World Congress, Econometric Society Monographs. Cambridge: Cambridge University Press.
Fehr, E., & Zehnder, C. (2006). Reputation and credit market formation. FINRISK Working Paper, University of Zurich.
Ferri, G., Kalmi, P., & Kerola, E. (2010). Organizational structure and performance in European banks: A reassessment. Paper prepared for the EURICSE Conference “Financial Co-operative Approaches to Local Development through Sustainable Innovation”.
GABV (Global Alliance for Banking on Value). (2012). Full report. http://www.gabv.org/wp-content/uploads/Full-Report-GABV-v9d.pdf.
Gächter, S., & Falk, A. (2002). Reputation and reciprocity: Consequences for the labour relation. Scandinavian Journal of Economics, 104, 1–27.
Global Report Initiative. (2011). Sustainability reporting guidelines, version 3.1. www.globalreporting.org.
Gouteroux, C. (2006). Le système bancaire et financier français en 2005. Bulletin de la Banque de France, 151, 75–85.
Green, C. F. (1989). Business ethics in banking. Journal of Business Ethics, 8, 631–634.
Grunert, J., Norden, L., & Weber, M. (2005). The role of non-financial factors in internal credit ratings. Journal of Banking & Finance, 29, 509–531.
Gutiérrez-Nieto, B., Serrano-Cinca, C., & Camón-Cala, J. (2011). A credit score system for socially responsible lending. CEB Working Paper No 11/028. Université Libre de Bruxelles.
Hudon, M. (2007). Fair interest rates when lending to the poor. Ethics and Economics, 5, 1–8.
Iannotta, G., Nocera, G., & Sironi, A. (2007). Ownership structure, risk and performance in the European banking industry. Journal of Banking & Finance, 31, 2127–2149.
Jaffee, D. M., & Russell, T. (1976). Imperfect information, uncertainty, and credit rationing. Quarterly Journal of Economics, 90, 651–666.
Karlan, D. (2005). Using experimental economics to measure social capital and predict financial decisions. American Economic Review, 95, 1688–1699.
Kitson, A. (1996). Taking the pulse: Ethics and the British cooperative bank. Journal of Business Ethics, 15, 1021–1031.
La Nef. (2001, 2002, 2003, 2004, 2006, 2010). Annual reports.
Liberti, J., & Mian, A. (2009). Estimating the effect of hierarchies on information use. Review of Financial Studies, 22, 4057–4090.
Machauer, A., & Weber, M. (1998). Bank behavior based on internal credit ratings of borrowers. Journal of Banking & Finance, 22, 1355–1383.
McLeish, K. J., & Oxoby, R. J. (2011). Social interactions and the salience of social identity. Journal of Economic Psychology, 32, 172–178.
Norman, W., & MacDonald, C. (2004). Getting to the bottom of ‘triple bottom line’. Business Ethics Quarterly, 14, 243–262.
Ory, J. -N., Jaeger, M., & Gurtner, E. (2006). La banque à forme coopérative peut-elle soutenir durablement la compétition avec la banque SA? Finance Contrôle Stratégie, 9, 121–157.
Périlleux, A., Hudon, M., & Bloy, E. (2012). Surplus distribution in microfinance: Differences among cooperatives, nonprofit, and shareholder forms of ownership. Nonprofit and Voluntary Sector Quarterly, 41, 386–404.
Petersen, M., & Rajan, R. (1994). The benefits of lending relationships: Evidence from small business data. Journal of Finance, 49, 3–37.
Rabin, M. (1993). Incorporating fairness into game theory and economics. American Economic Review, 83, 1281–1302.
Robert de Massy, O., & Lhomme, G. (2008). Les nouvelles frontières de la formation bancaire. Revue d’Economie Financière, 92, 229–243.
San-Jose, L., Retolaza, J. L., & Gutierrez, J. (2011). Are ethical banks different? a comparative analysis using the radical affinity index. Journal of Business Ethics, 100, 151–173.
Scholtens, B. (2006). Finance as a driver of corporate social responsibility. Journal of Business Ethics, 68, 19–31.
Stiglitz, J., & Weiss, A. (1981). Credit rationing in markets with imperfect information. American Economic Review, 71, 393–410.
Taupin, M. T., & Glémain, P. (2007). Les logiques d’acteurs des finances solidaires contemporaines: Entre innovation et résilience. Annals of Public and Cooperative Economics, 78, 629–661.
Uzzi, B. (1999). Embeddedness in the making of financial capital: how social relations and networks benefit firms seeking financing. American Sociological Review, 64, 481–505.
Weber, O., & Remer, S. (2011). Social banks and the future of sustainable finance. London: Routledge.
Acknowledgments
The authors thank Yiorgos Alexopoulos, Francesca Barigozzi, Régis Blazy, Carlo Borzaga, Damien Brousolle, Isabelle Cadoret, Anastasia Cozarenco, Jacques Defourny, Joeffrey Drouard, Silvio Goglio, Marek Hudon, Marc Jegers, Panu Kalmi, Georg Kirchsteiger, Philipp Koziol, Marc Labie, Neil McHugh, Fabien Moizeau, Jonathan Morduch, Tomasso Oliveiro, Anaïs Périlleux, Jose Luis Retolaza, Michael Roberts, Leire San-Jose, Jessica Schicks, Hubert Tchakoute Tchuigoua, Piero Tedeschi, Gregory Udell, Olaf Weber, Laurent Weill, the participants at the CERMi Seminar, ULB (May 2012), the “Cooperative Finance and Sustainable Development” Conference at the University of Trento (June 2012), the “Frontiers of Finance” Workshop at the Paris Panthéon Sorbonne University (October 2012), the Workshop on SME Finance at the University of Strasbourg (April 2013), the Third European Research Conference on Microfinance at the University of Agder (June 2013), the “Finance and Society” Workshop at BEM/KEDGE Business School (June 2013), the EMES Conference in Liège (July 2013), as well as an anonymous referee for helpful comments and discussions. This research has been carried out in the framework of an “Interuniversity Attraction Pole” on social enterprise, funded by the Belgian Science Policy Office.
Author information
Authors and Affiliations
Corresponding author
Appendices
Appendix 1: La Nef’s Organizational Characteristics
See Table 9.
Appendix 2: Technicalities in the Cost-Benefit Analysis of Reciprocity
Here we report the detailed computation of three components of NBR in Eq. (3).
Computation of ∆CD t
La Nef is committed to report loan-disaggregated LLPs to the French banking authority on a quarterly basis. We managed to gain access to the report released in the first quarter of 2007 (this also gives the level of provisioning for the last quarter of 2006), while our sample period ends in November 2008. As a result, we have detailed information on LLPs for 65 loans out of the 91 defaulted loans in our sample (i.e., 71.4 %). We have estimated the missing LLPs by multiplying the respective loan sizes by the average provisioning rate computed from the observable LLPs. This average rate is 27.54 %.Footnote 33 One could object that LLPs are adjusted over time in reaction to changes in default expectations. In practice, however, the adjustments prove to be limited. Between the last quarter of 2006 and the first quarter of 2007, the average LLP adjustment was 2.58 % only. Therefore, we consider that the missing 1-year adjustment does not affect ∆DC much. Last, we discounted all the LLPs according to the year of default.
To measure how reciprocity reduces the cost of default, we run a Tobit regression (see Table 10). The explained variable is the present value of LLPs for defaulted loans, and 0 otherwise. The explanatory variable of interest is SR. We also include control variables accounting for contractual features, financial risk characteristics, and relational aspects. The marginal effects reported in Column (2) indicate that the present value of LLP decreases by €1,047.72 per unit of SR. In this way, we obtain the differential LLPs driven by each actual loan with SR = 2 or 3. Summing up, we obtain an estimate of the total benefit attributable to the reduction in yearly default occurrences.
Computation of ∆I t
We compute the differential in cashed-in interests as follows. For each loan in our sample, we compare two situations: The actual one and its “SR = 1” simulated counterpart. The aim is to compute the discounted cashed-in interests for the two situations, and then take the difference between them. To simplify the computations, we work out annual installment (constant annuities) even though borrowers repay in monthly installments.
Table 11 depicts an example. The 5-year loan amounts €50,000. The actual interest charged by the bank on this loan is 5 % and the actual SR is 3. From Table 5 (Specification (3)), we find that the simulated counterpart of the loan bears interest at 5.30 %. Table 5 extracts the annual interest paid on both loans (5 and 5.30 %). Annual differences are then computed and discounted at a 6 % rate. The final result is the sum of these figures, i.e., €418.05. Similar calculations are conducted for all the loans in our sample.
Computation of SC t
Loan officers represent the main cost drivers of screening costs. To evaluate the proportion of the extra cost dedicated to social screening, we gauge the productivity of La Nef’s loan officers compared with that of loan officers in non-social banks dealing with the same type of borrowers (i.e., small- and medium-sized enterprises, henceforth SMEs). The productivity of a loan officer is proxied by the number of loans she grants annually. Informal contacts with La Nef’s managers have revealed that, according to their standard, a full-time loan officer grants 25 loans annually. In comparable non-social banks, we have estimated this load to be 36.67.Footnote 34 We therefore attribute 31.84 % of the workload of loan officers in La Nef to social screening.
We use the conservative assumption that the screening operation overheads (SCO) are fully captured by the operational costs associated with loan officers, including wages. To determine those costs, we proceed as follows (see Table 12). First, we extract from La Nef’s annual reports the yearly overhead expenses incurred by all the bank’s operations, the yearly full staff sizes, and the yearly numbers of active loan officers (La Nef 2001, 2002, 2003, 2004). Second, we compute the year-t average cost per staff member by dividing the overhead expenses in year t by the number of full-time staff members active during year t. Third, we derive the year-t SCO by multiplying the number of full-time loan officers active in year t by the year-t average cost per staff member. The SCOs include both financial and social screening costs but exclude those associated with back-office personnel. Last, to estimate \(SC_{t}\), we multiply the year-t SCO by the 31.84 % factor representing the excess workload of loan officers due to social screening.
Appendix 3: Additional Robustness Checks
We carry out four additional robustness checks on the full sample. First, Table 12 gives the results from the multivariate estimation of specification (2) in Table 5. It is based on reduced-form estimation. In this way, we assess the impacts of loan characteristics on credit conditions, while avoiding potential endogeneity biases.
Overall, the figures in the first column of Table 13 confirm those in Table 5 regarding the impact of the social rating on interest rates. R 2 indicates that the adjustment is poor for the other credit conditions. This can be attributed to two factors. First, loan size alone is a loose indicator of credit rationing. Its determination is most likely influenced by the requested amount, which is unobservable. Second, collateralization for start-ups is highly dependent on public guarantees, which are also unobservable. Both limitations might create missing-variable distortions in the estimations of loan size and collateralization. Nevertheless, reduced-form estimation has the merit of freeing the interest rate loadings from these distortions.
Second, in Table 14 we propose two alternative specifications for the measurement of the social rating. First, we use two-step estimations to clean SR of its interactions with other loan characteristics. In column (1), an ordered probit regression model extracts the residuals of SR when regressed on FIN, STARTUP, and RELATIONSHIP. These residuals constitute “pure” SRs (PSR). Only the start-up dummy is significant in the first-path regression. Then, the interest rate (column (2)) is estimated by substituting PSR for SR. The empirical results prove to be robust to this change. The negative impact of PSR on the probability of default (column (3)) is the same as that of SR in Table 6. We thus exclude any spurious effect due to accidental correlations between SR and other loan characteristics. Second, in columns (4) and (5), we dichotomize SR and use variable DICSR, which takes value 0 if SR = 1, and 1 when SR > 1. The aim is to limit the impact of the ordinality of SR. According the La Nef’s criterion, DICSR = 1 indicates that the projects have at least one social or environmental component. Except for the significance level, which passes from 1 to 5 %, dichotomizing SR does not modify the previous results. The negative impact of DICSR on the probability of default (column (5)) is even stronger than that of SR in Table 6. Altogether, Table 4 not only confirms our previous results, they also emphasize that our findings are driven by purely social motives.
Third, we run instrumental-variable estimation to account for the possibility of SR being endogenous. The results (not reported here) show that the Durbin–Wu–Hausman test fails to reject the null hypothesis that SR is exogenous (p = 0.6584).Footnote 35
Fourth, Table 15 proposes two specifications including additional explanatory variables. Columns (1) and (2) in Table 15 examine whether the impact of SR is partly attributable to loan size. The descriptive statistics in Sect. 3 pointed out that the borrowers with higher social ratings tend to receive larger loans. In fact, we checked the potential effect of loan size in two ways. First, we estimated the two equations (for rate and default) on a censored sample obtained by excluding the largest loans. Several cut-off points were used (results not reported). All of them produced results consistent with those from our baseline regressions. Second, we added the interaction between loan size and SR among the explanatory variables. Table 15 reveals that the loadings of this interaction term in our two regressions of interest are insignificant. Loan size does not interfere with the reciprocity effect.
As explained in Sect. 3, some loans, especially those made to start-ups, benefit from public collateral. The subsequent incentive may affect the bank’s lending behavior. We investigate this possibility in columns (3) and (4) in Table 15 by including the dummy variable PUBLIC COLLAT (equal to 1 if the loan benefits from public collateral, and 0 otherwise). The regression results show that public collateral has a significantly negative impact on the probability of default. Meanwhile, the STARTUP dummy loses significance, which might indicate the presence of multicollinearity between STARTUP and PUBLIC COLLAT. In any case, the impacts of our variables of interest, SR and FIN, remain consistent with those obtained from our baseline regressions.
Rights and permissions
About this article
Cite this article
Cornée, S., Szafarz, A. Vive la Différence: Social Banks and Reciprocity in the Credit Market. J Bus Ethics 125, 361–380 (2014). https://doi.org/10.1007/s10551-013-1922-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10551-013-1922-9