
































Microsoft Word - APE-V2N1-p23


Advances in Politics and Economic 
ISSN 2576-1382 (Print) ISSN 2576-1390 (Online) 

Vol. 2, No. 1, 2019 
www.scholink.org/ojs/index.php/ape 

23 
 

Original Paper 

Economic Incentives and Social Norms Can Lead Actors to 

Prosocial Behavior: An Evidence-Based Framework and 

Findings 

Angbonon E. Kamalan1* 
1 Department of Economics, Alassane Ouattara University, Bouaké, Ivory Coast 
* Angbonon E. Kamalan, Department of Economics, Alassane Ouattara University, Bouaké, Ivory 

Coast 

 

Received: February 1, 2019   Accepted: February 16, 2019    Online Published: February 18, 2019 

doi:10.22158/ape.v2n1p23              URL: http://dx.doi.org/10.22158/ape.v2n1p23 

 

Abstract 

This article attempts a search for economic, sociological and even psychological answers about the 

foundations of individual’s prosocial behavior with a focus on economic incentives and social norms. 

First, we present an evidence-based approach of individual’s prosocial behavior using the justification 

and the economies of worth of Boltanski and Thevenot (2006). After that, we analyze a prosocial 

behavior applied to the process of maturity repayment of individual loans. We propose two different 

methods to check the theoretical propositions: logit and probit method in one hand and fractional 

method of Papke and Wooldridge (1996) in other hand. The logit and probit method are estimated in 

the aim of better understanding of the influencing factors of the maturity repayment behavior of 

borrowers. Estimation of fractional method helps to learn about the probability of delinquency in loan 

repayment. The results attest the theoretical propositions. 

Keywords 

incentives, social norms, prosocial behavior, probit, fractional 

 

1. Introduction 

Both economic incentives and social norms influence individual behavior. Many authors including 

Nobel Prizes have shed light on this matter by providing excellent works specifically relating to 

behavioral economics (Note 1). The rational choice school of sociologists also recognizes the joint 

influence of social norms and economic incentives (Lindbeck et al., 1999). While economists have 

focused on economic incentives, sociologists have emphasized social norms. 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

24 
Published by SCHOLINK INC. 

Prosocial activities represent a substantial part of social life, including such actions as donating money 

for a cause or an organization, volunteering for a party during election times, voting, cleaning beaches, 

donating blood, etc. (Lacetera & Macis, 2010).  

Understanding what motivates individuals to contribute to prosocial activities emerges, therefore, as a 

topic of increasing interest in economics as well as sociology and even psychology. The issue is made 

all the more pressing by the fact that, for many of these activities, supply is often below societal needs.  

However, individuals might simply not find it worthwhile to engage in prosocial activities if the 

benefits fall short of the opportunity costs. If this is the case, then material incentives (composed of 

material cash-incentives also called explicit economic incentives, and material non-cash incentives) 

might be effective in increasing the number and frequency of prosocial acts. But, a large common idea 

is that, explicit economic incentives designed to increase contributions to public goods and to promote 

other prosocial behavior sometimes are counterproductive or less effective than would be predicted 

among entirely self-interested individuals. This may occur when incentives adversely affect 

individuals’ altruism, ethical norms and intrinsic motives (Bowles & Polanía-Reyes, 2012).  

Bénabou and Tirole (2006) enlightened this issue in their study of incentives and prosocial behavior 

and provided convincing answers which defined extrinsic incentives given by the society and the 

intrinsic incentives providing from personal motivation. They develop a theory of prosocial behavior 

that combines heterogeneity in individual altruism and greed with concerns for social reputation or 

self-respect.  

To understand what kind of incentives might encourage prosocial behavior, however, one must first 

have an understanding of the motives behind altruistic behavior. In fact, recent empirical and 

theoretical contributions suggest that depending on what motivates individuals to contribute to 

prosocial causes, certain types of material incentives might backfire (Frey & Oberholzer-Gee, 1997; 

Ryan & Deci, 2000; Lacetera & Macis, 2010).  

In addition, the effect of economic incentives and social norms in shaping of prosocial behaviors of 

individuals is explained in a sociological approach based on the theory of reciprocity that defines 

reciprocal behavior as a social preference (Simpson & Willer, 2008).  

Although there are several types of social preferences in the literature, Fehr and Fischbacher (2002) 

provide the most common characteristics of social preferences. Particularly important types of social 

preferences are the reciprocal fairness (not driven by the expectation of future material benefit), the 

inequity aversion, pure and impure altruism; and self-identity preferences. In short, the sociological 

approach of prosocial behavior focus on three sets of theories addressing non‐selfish motives or 

“other‐regarding” behavior, namely: (1) the outcome‐based prosocial preferences theories (inequity 

aversion, pure and impure altruism) that assume that an individual’s utility depends directly on the 

utility of other people; (2) the theories of reciprocity (or conditional cooperation) that are based on the 

notion that individuals behave in a friendly manner when they are treated benevolently and, conversely, 

they act meanly when treated badly; and (3) the approaches stressing the importance of self‐identity for 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

25 
Published by SCHOLINK INC. 

prosocial behavior. These theoretical frameworks are relevantly interpreted by Boltanski and Thevenot 

(2006) for better understand how actors can justify their behaviors using contradictory foundations. 

This article requires a search for economic, sociological and even psychological answers about the 

foundations of individual’s prosocial behavior with a focus on economic incentives and social norms. 

So the study combines economic theories and sociologic approach of prosocial behavior using field 

experiment. An illustration is provided using the maturity repayment behavior of loans in a specific 

context.  

We propose a socioeconomic framework that clearly introduce to the understanding of the borrower’s 

prosocial behavior regarding maturity repayment. Then, an empirical illustration of the theoretical 

propositions is given, based on a survey data. 

The article is organized as follows. Section 2 presents an evidence-based approach of individual’s 

prosocial behavior based on the justification and the economies of worth of Boltanski and Thevenot. 

Section 3 proposes the empirical evidence with different methods (logit, probit and fractional) used to 

check the theoretical propositions and provides results interpretation. Finally, section 4 concludes. 

 

2. Evidence-Based Framework of Prosocial Behavior 

2.1 Context and Issues 

The study applies to the context in which a debt relationship exists between a lending institution 

represented by a credit manager and a borrower. The prosocial behavior being studied is the maturity 

repayment behavior of credit in microfinance institutions. Indeed, in Microfinance Institutions (MFIs) 

in West Africa, lending contracts are often drawn up taking into account the borrower’s relationships 

with third parties, whether the borrower is in liability group lending or whether he is in individual 

lending (Besley, 1995). Individual lending is most concerned with endorsement and sponsorship 

relationships, which are known as non-market institutions applied for loans, Figure 1. 

 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

26 
Published by SCHOLINK INC. 

 

Figure 1. Relationship within Individual Lending 

 

The endorsement and sponsorship practices are formalized as supervisory mechanism for better project 

selecting and monitoring of borrowers within lending contracts (Kamalan, 2018). The endorsement 

practice refers to relationships between an individual borrower who does not have personal material 

collateral and the third party, the endorser. The latter is usually a person close to the borrower, for 

example the spouse or member of the family lineage. The endorser provides a material asset as a 

collateral in return to the debt the borrower receives. The endorsement relationship engages the 

endorser to have the same level of commitment as the joint liability relationship.  

The sponsorship is similar to the moral assistance that generally involves a testimonial process. It 

consists in the fact that, the third party close relative of the borrower, testifies the good probity and 

exemplary morality of the borrower in order to dispel any doubt that may remain with a credit manager 

(in a lending institution) when examining the borrower’s loan application. In contrast to the endorser, 

the sponsor is simply a moral guarantor who commits his moral honor for the benefit of the borrower 

by testifying in his behalf. The sponsor makes himself available to the lending institution for assistance 

in recovering the assets, as long as the borrower does not fulfil his commitments.  

If the endorser or the sponsor is not a borrower himself, he does not derive any direct benefit from his 

act; that is to be considered as a pure altruism. The main interest of endorser or sponsor is to assist the 

borrower in the credit application. Endorser and sponsor may be satisfied with offering an opportunity 

to a family member for whom they feel morally indebted because of first-born obligation for example.  

In several cases, borrowers engaged in relationships with third parties such as joint borrowers in 

liability group have been known for having better loan repayment behavior (Armendáriz-de-Aghion & 

Gollier, 2000). This behavior for loan repayment is all time explained by the social pressure of the 

group members who want the credit to be granted repeatedly to each one in turn. But, the social 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

27 
Published by SCHOLINK INC. 

pressure is less effective and often inactive or counterproductive when the borrower has relationships 

with third parties who are not engaged in the same lending or who are outside the lending institution. 

This is the case for sponsorship and endorsement relationships. In this case, the design of lending 

contract and the repayment behavior must be analyzed using another prism: the prosocial behavior. 

Answering the questions raised by the use of sponsorship and endorsement relationships within lending 

relationships, an interesting approach is provided by the theory of social capital originally defended by 

P. Bourdieu, J-C. Coleman and R. Putnam. Putnam’s approach, in contrast to the Bourdieusian ones, 

provides a framework of social capital that makes social ties an essential endowment of a society. This 

feature makes resources available to members to be active within the functioning of the society. Putnam 

believes that a society can potentially have a significant endowment of social capital in the form of 

trust, reciprocity norms and civic engagement networks, for to facilitate voluntary cooperation between 

individuals in the sense of collaboration, knowledge pooling, skills and efforts sharing. This approach 

of social capital is closely in line with the early work of institutional economics. 

Therefore, sponsorship and endorsement relationships are considered as societal attributes rather than 

individual assets. A problem is to effectively engage the attributes of a society and make them interact 

in the meaningful behavior of an individual, as observed in the lending relationship. It is clear that the 

individual’s aim is to maximize earnings and expectations. Consequently, it will be asked to determine 

the process to be use to resolve conflicts between the collective aspirations of society induced by 

societal attributes and the rational expectations of individuals motivated by selfish interests. These 

issues are strongly addressed by Boltanski and Thevenot (2006) in their analysis of justification and the 

economies of worth. 

2.2 Evidence-Based Practices that Shape Behavior 

To better understanding the sponsorship and endorsement relationships within lending contracts and the 

connection to the economies of worth developed by Boltanski and Thevenot, we consider the sense of 

credit and the effects on behavior shaping. That opens up a normative dimension of the reasoning about 

the meaning of the lending for the actors involved, as stakeholders of the lending contract; particularly 

for the borrower. That’s the justification of the behavior based on the foundations basically called 

“values”, Kamalan (2019). We’ve matched the economies of worth of Boltanski and Thevenot with the 

values highlighted by the sponsorship and endorsement relationships, Figure 2.  

 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

28 
Published by SCHOLINK INC. 

 

Figure 2. Foundations of Values Highlighted by Sponsorship and Endorsement 

 

Let’s only consider the cities of market, fame and inspiration defined by Boltanski and Thevenot (Note 

2). Boltanski and Thevenot defined the city of “market” that uses competition as a greatness. This city 

highlights the personal advantages of the actor (the borrower) whose interests are freed from domestic 

and civic considerations. The link is no longer synonymous with identity filiation, but refers to a 

consumerist attachment. In this context, the borrower’s desirable behavior is to be sought in the 

self-determination and self-sacrifice in preserving the financial relationship with the lending institution, 

no matter the costs and effort to be expended. Value takes on a sense of competence. Borrower is 

valuable because he or she is capable. The value of competency is all the more sought if it is rewarded. 

The credit renewal willingness with higher amount is one of the best tool used to test the impact of 

competency on maturity repayment behavior. That is defined as an economic material incentive. 

In the city of “industry”, greatness is revealed in a form of the individual’s effectiveness in overcoming 

the challenges faced by all individuals. Value is based on the individual’s success in overcoming these 

challenges. That value of success is all the more sought if it is rewarded. Lot of variables can be used to 

test the impact of the value of success in the behavior of maturity repayment: indifference regarding 

failure, appreciation of post-contractual opportunism, etc. There are defined as economic non material 

incentives. 

In the city of “fame”, greatness is reflected in the opinion that includes self-esteem and the reputation. 

That’s the positive reputation or the looking-glass-self. An example is the fact that the more a borrower 

is motivated by self-esteem, the more he behaves in actions that are admired by those surrounding him 

and the higher is the maturity repayment. Finally, in the city of “inspiration”, value refers to the 

religious notion of beliefs. Individuals have a moral commitment to comply with the high standards of 

a virtuous lifestyle as required by religious norms.  

Some variables such us happiness/jealousy of neighborhoods, past shame experiences, etc. can be used 

to test the impact of the value of fame in the behavior of maturity repayment and marital or religious 

status are supposed to have benefic impact on the behavior of maturity repayment. All these variables 

are regarding social norms. 

The next section proposes the empirical evidences in which theoretical analyses are illustrated with 

data from a field study in a West-African microfinance institution (PAMDE) in Benin (Kamalan, 2010).  



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

29 
Published by SCHOLINK INC. 

3. Empirical Evidence 

3.1 Data, Variables and Methods 

In that section, we’ll focused on a survey covered sample of 832 customers composed of 506 individual 

loans and 272 group loans, representative the population of 21146 in a microfinance institution called 

PADME in Cotonou (Benin). The survey includes only borrowers in process of credit. The survey was 

conducted in 2006 while PADME was considered to be one of the best microfinance institutions with 

higher repayment rate of borrowers. PADME has been selected because it gives opportunity to learn 

about the non-market institutions such as endorsement and sponsorship practices involved in lending 

contracts.  

The methods used in that section are the logit and probit models and the fractional regression model of 

Papke and Wooldridge (1996). The logit and probit models are used to study the behavior for loan 

repayment. The variable of interest is the “delays” in loan maturity repayment which takes the value 1 

if the borrower has no default in maturity repayment and takes the value 0 if at least one maturity 

repayment is defected. The explanatory variables are displayed in Figure 3. 

 

 

Figure 3. Explanatory Variables 

 

The fractional model helps to understand the probability of delinquency in loan repayment. This model 

is useful for situations where the variable of interest is continuous and restricted to the interval (0, 1) 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

30 
Published by SCHOLINK INC. 

and is related to other variables through a regression structure (Cribari-Neto & Zeileis, 2010). With the 

fractional method, we are interested in knowing about the behavior of the delay in repayment. It seems 

like delinquency in maturity repayment. First we generate the P probability corresponding to the 

situation where there is no delay. Then, we create the variable R=1-P corresponding to the probability 

of having at least 1 delay. We generate the R variable and we estimate the fractional regression model. 

That model is generally used for beta regression where variables are greater than zero and less than one, 

such as rates, proportions and indices.  

3.2 Findings 

3.2.1 The Influencing Factors of the Maturity Repayment Behavior 

The Logit model estimate (Figure 4) shows that the factors that explain the maturity performance are: 

the indifference regarding failure, the marital status, the sponsorship, the willingness to renew the credit 

received, status and type of loan. Only significant variables are considered. These variables are 

supposed to have positive effects on repaying maturities without delay, in contrast to the type of loan 

and the marital status which have negative effects.  

The Probit model (Figure 5) is more restrictive since marital status is not significant.  

 

                                                                              
       _cons    -3.733109    1.24912    -2.99   0.003     -6.18134   -1.284878
      famsup    -.0667715   .1809356    -0.37   0.712    -.4213988    .2878558
     loantyp    -1.417239   .2424026    -5.85   0.000    -1.892339   -.9421381
     postopp     .0836893   .1841911     0.45   0.650    -.2773186    .4446973
    loanrenw     .6551918   .3318552     1.97   0.048     .0047677    1.305616
     sponsor     2.597394    .205262    12.65   0.000     2.195088    2.999701
    relistat     1.483343   1.136598     1.31   0.192    -.7443474    3.711033
    maristat    -.5697063   .2987465    -1.91   0.057    -1.155239    .0158261
     shamexp     .0664903   .2095035     0.32   0.751    -.3441291    .4771097
    indifail     .7921605   .2285467     3.47   0.001     .3442172    1.240104
    jealneig     .1379401   .3114363     0.44   0.658    -.4724639    .7483441
     hapneig      .555103   .4105596     1.35   0.176    -.2495791    1.359785
     endorse     .2079762   .2155437     0.96   0.335    -.2144817     .630434
                                                                              
      delays        Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
                                                                              

Log likelihood = -422.92268                     Pseudo R2         =     0.2665
                                                Prob > chi2       =     0.0000
                                                LR chi2(12)       =     307.24
Logistic regression                             Number of obs     =        832

 

Figure 4. Logistic Regression 

 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

31 
Published by SCHOLINK INC. 

                                                                              
       _cons    -2.322556   .7149657    -3.25   0.001    -3.723863   -.9212487
      famsup    -.0522041    .106517    -0.49   0.624    -.2609736    .1565655
     loantyp    -.8213902   .1377351    -5.96   0.000    -1.091346   -.5514343
     postopp     .0472681   .1081512     0.44   0.662    -.1647044    .2592406
    loanrenw     .3775101   .1964478     1.92   0.055    -.0075205    .7625408
     sponsor     1.560659   .1157665    13.48   0.000     1.333761    1.787557
    relistat     .9220995   .6420364     1.44   0.151    -.3362688    2.180468
    maristat    -.2650423    .166067    -1.60   0.110    -.5905277    .0604431
     shamexp     .0397269    .123447     0.32   0.748    -.2022247    .2816785
    indifail     .4480144   .1351276     3.32   0.001     .1831691    .7128596
    jealneig     .0733384   .1786995     0.41   0.682    -.2769062     .423583
     hapneig     .3203205   .2428238     1.32   0.187    -.1556054    .7962463
     endorse     .1392684   .1313471     1.06   0.289    -.1181672    .3967041
                                                                              
      delays        Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
                                                                              

Log likelihood = -422.60067                     Pseudo R2         =     0.2670
                                                Prob > chi2       =     0.0000
                                                LR chi2(12)       =     307.89
Probit regression                               Number of obs     =        832

 

Figure 5. Probit Regression 

 

Appendix 1 and Appendix 3 provide more details for the Logit and Probit models.  

With regard to the marginal effects after logit and probit regression, the following results can be noted: 

the probability of having no delay increases by 18.92% (Appendix 2) and 17.24% (Appendix 4) when 

moving from a borrower which is indifferent regarding failure to another one who is very embarrassed. 

Religious status concerning non-current practicing to current practicing borrower increases the 

probability of having no delay by about 30%.  

The willingness of loan renewal is also powerful in increasing the probability of having no delay by 

about 14%.  

The most powerful tool that lead borrowers to increasing the probability of having no delay is the 

sponsorship by 54.82% when moving from one borrower without sponsor to another one who is 

sponsored. Finally, we note that probability of not being late decreases by 31.84% when moving from 

joint liability group loan to individual loan. That suppose that joint liability group loans are less risky 

than individual loans. 

3.2.2 The Probability of Delinquency in Repayment Behavior 

Several factors explain the probability of having at least one delay in repayment (Figure 6). 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

32 
Published by SCHOLINK INC. 

                                                                              
       _cons      .349178   .1427209     2.45   0.014     .0694503    .6289058
      famsup     .1827017   .0518445     3.52   0.000     .0810883    .2843151
     postopp     .0068026    .053586     0.13   0.899    -.0982241    .1118293
    maristat     .2484953   .0877599     2.83   0.005     .0764891    .4205015
     shamexp      .222241   .0572201     3.88   0.000     .1100916    .3343904
    jealneig    -.3493436   .0851817    -4.10   0.000    -.5162966   -.1823905
     hapneig    -.4144738   .0913214    -4.54   0.000    -.5934605   -.2354871
     endorse     .0247069   .0489988     0.50   0.614     -.071329    .1207428
                                                                              
           R        Coef.   Std. Err.      z    P>|z|     [95% Conf. Interval]
                             Robust
                                                                              

Log pseudolikelihood =  -565.1106               Pseudo R2         =     0.0192
                                                Prob > chi2       =     0.0000
                                                Wald chi2(7)      =      83.48
Fractional probit regression                    Number of obs     =        832

 

Figure 6. Fractional Regression 

 

The interpretations based on the conditional mean of R predicted in Appendix 5 give the conclusions 

above. The probability of having at least one delay decreases on average about 15.98% when moving 

from a borrower who consider the neighborhood not being happy of his or her financial situation to 

another borrower whose neighborhood is qualified as being happy. The reasoning is the same with the 

jealousy neighborhood variable (13.6% reduction). For the variables (shame experience) and (marital 

status), the probability of having at least one delay increases about 8% or 9% when passing from a 

borrower who has never experienced a shameful situation to another who has already experienced a 

shameful situation; or from a not married to a married borrower. Finally, moving from borrower who 

never ask for family financial support to another who often ask for family financial support increases 

about 7.2% the probability of having at least one delay in maturity repayments. 

 

4. Concluding Remarks 

Individual decisions concerning prosocial activates belong to the class of decisions where both 

economic incentives and social norms play a role. In this paper, we attempted to bring together social 

norms and economic incentives on an equal footing in a model of individual choice concerning 

prosocial behavior, especially related to maturity repayment. After describing an overview of 

economics and sociological theories that address the motives of individual’s prosocial behavior, we 

proposed an evidence-based sociological framework of individual’s prosocial behavior focusing on the 

justification and the economies of worth of Boltanski and Thevenot.  

This article explores the frontier of economy and sociology by questioning the very foundations of 

individual’s prosocial behavior. The study addresses the causal effect of economic incentives and social 

norms on individual moral hazard applied to maturity repayment behavior of loans. Without using 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

33 
Published by SCHOLINK INC. 

collateral as defined in the microeconomics of banking, we studied theoretically the effects of values 

used as the very bases of the reasoning that justify individual behaviors in various contexts. And we 

empirically tested the effects of theses values in the maturity repayment behaviors of borrowers in a 

specific context applied in Africa. We used the Logit, the Probit and the Fractional method.  

These results invite to a socioeconomic approach which indicates that economic material and 

non-material incentives and social norms known as social preferences are likely to have important 

effects on shaping of prosocial behaviors of individuals. The results indicate that the probability of 

having no delay strongly increase with the use of the practice of sponsorship. We notice that the 

practice of endorsement is non-significant for knowing about the behavior in maturity repayment. That 

help us conclude that social norms (such as religious status, indifference regarding failure) and material 

incentives (loan renewal) and non-material incentives (sponsorship) are potentially powerful to lead 

borrowers to the prosocial act of maturity repayment of loans.  

However, the logit and probit models explain only 26% of borrowers’ behaviour towards maturity 

repayments. This recommends including additional explanatory variables in our models or looking for 

alternative models such as Poisson regression analysis. 

 

References 

Armendáriz-de-Aghion, B., & Gollier, C. (2000). Peer Group Formation in an Adverse Selection Model. 

The Economic Journal, 110(465), 632-643. https://doi.org/10.1111/1468-0297.00557 

Bénabou, R., & Tirole, J. (2006). Incentives and Prosocial Behavior. American Economic Review, 96(5), 

1652-1678. https://doi.org/10.1257/aer.96.5.1652 

Besley, T. (1995). Non market Institutions for credit and risk sharing in low-Income. Journal of 

Economic Perspectives, 9(3), 115-127. https://doi.org/10.1257/jep.9.3.115 

Boltanski, L., & Thevenot, L. (2006). On Justification, Economies of Worth. Catherine Porter, French 

initial version: De la justification, les économies de la grandeur, Gallimard. 

Bowles, S., & Polanía-Reyes, S. (2012). Economic Incentives and Social Preferences: Substitutes or 

Complements? Journal of Economic Literature, 50(2), 368-425. 

https://doi.org/10.1257/jel.50.2.368 

Cribari-Neto, F., & Zeileis, A. (2010). Beta Regression in R. Journal of Statistical Software, 34(10), 

1-24. 

Fehr, E., & Fischbacher, U. (2002). Why social preferences matter-the impact of non-selfish motives on 

competition, cooperation and incentives. The Economic Journal, 112(478), 1-33. 

https://doi.org/10.1111/1468-0297.00027 

Frey, B., & Oberholzer-Gee, F. (1997). The cost of price incentives: An empirical analysis of 

motivation crowding-out. American Economic Review, 87(4), 746-755. 

Kamalan, A. E. (2010). Efficacité des contrats de dette pour les clients individuels en microfinance, le 

rôle des liens de clientèle, étude dans des institutions de microfinance oust-africaines. Lyon, Thèse 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

34 
Published by SCHOLINK INC. 

de doctorat en science économique. Université Lyon, 2. 

Kamalan, A. E. (2018). Supervisory incentives within contracts: A principal-supervisor-agent approach. 

Journal of Finance and Bank Management, 6(2), 58-63. 

Kamalan, A. E. (2019). Economics and Sociological Foundations of Prosocial Behavior: A Theoretical 

Survey. American Journal of Humanities and Social Sciences Research, 3(1), 164-168. 

Lacetera, N., & Macis, M. (2010). Social image concerns and prosocial behavior: Field evidence from 

a nonlinear incentive scheme. Journal of Economic Behavior & Organization, 76(2010), 225-237. 

https://doi.org/10.1016/j.jebo.2010.08.007 

Lindbeck, A., Nyberg, S., & Weibull, J. (1999). Social norms and economic incentives in the welfare 

State. The Quarterly Journal of Economics, 114(1), 1-35. 

https://doi.org/10.1162/003355399555936 

Papke, L., & Wooldridge, J. (1996). Econometric methods for fractional response variables with an 

application to 401(K) plan participation rates. Journal of Applied Econometrics, 11, 619-632. 

https://doi.org/10.1002/(SICI)1099-1255(199611)11:6<619::AID-JAE418>3.0.CO;2-1 

Ryan, R., & Deci, E. (2000). Intrinsic and Extrinsic Motivations: Classic Definitions and New 

Directions. Contemporary Educational Psychology, 25(2000), 54-67. 

https://doi.org/10.1006/ceps.1999.1020 

Simpson, B., & Willer, R. (2008). Altruism and Indirect Reciprocity: The Interaction of Person and 

Situation in Prosocial Behavior. Social Psychology Quarterly, 71(1), 37-52. 

https://doi.org/10.1177/019027250807100106 

 

Notes 

Note 1. R. Thaler (2017, rationality and social preferences), R. Shiller (2013, Behavioral Finance), D. 

Kahneman and V. Smith (2002, cognitive bias, heuristic decisions, behavioral economics), H. Simon 

(1978, limited cognitive rationality). 

Note 2. Further details concerning the cities, greatness and values are provided in Boltanski and 

Thevenot (2006) and Kamalan (2010). 

 

 

 

 

 

 

 

 

 

 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

35 
Published by SCHOLINK INC. 

Appendix 1 

                  Prob > chi2 =         0.0000
            Pearson chi2(146) =       338.89
 number of covariate patterns =       159
       number of observations =       832

Logistic model for delays, goodness-of-fit test

 

                                                  
Correctly classified                        77.28%
                                                  
False - rate for classified -   Pr( D| -)   19.79%
False + rate for classified +   Pr(~D| +)   25.28%
False - rate for true D         Pr( -| D)   18.87%
False + rate for true ~D        Pr( +|~D)   26.42%
                                                  
Negative predictive value       Pr(~D| -)   80.21%
Positive predictive value       Pr( D| +)   74.72%
Specificity                     Pr( -|~D)   73.58%
Sensitivity                     Pr( +| D)   81.13%
                                                  
True D defined as delays != 0
Classified + if predicted Pr(D) >= .5

   Total           408           424           832
                                                  
     -              77           312           389
     +             331           112           443
                                                  
Classified           D            ~D         Total
                       True         

Logistic model for delays

 

Appendix 2 

(*) dy/dx is for discrete change of dummy variable from 0 to 1
                                                                              
  famsup*   -.0166407      .04511   -0.37   0.712  -.105054  .071772   .653846
 loantyp*   -.3390236      .05273   -6.43   0.000  -.442372 -.235675   .673077
 postopp*    .0208265      .04578    0.45   0.649  -.068906  .110559   .653846
loanrenw*    .1561585      .07361    2.12   0.034   .011893  .300424   .927885
 sponsor*    .5549829      .03232   17.17   0.000    .49163  .618336   .612981
relistat*    .3040212      .16111    1.89   0.059  -.011745  .619788   .990385
maristat*   -.1410714       .0721   -1.96   0.050  -.282388  .000245    .90024
 shamexp*    .0165782      .05227    0.32   0.751  -.085879  .119035   .216346
indifail*    .1892248      .05092    3.72   0.000   .089418  .289031   .813702
jealneig*    .0342094      .07681    0.45   0.656  -.116331  .184749   .894231
 hapneig*    .1332654      .09298    1.43   0.152  -.048978  .315509   .956731
 endorse*     .051812      .05364    0.97   0.334  -.053326   .15695    .41226
                                                                              
variable        dy/dx    Std. Err.     z    P>|z|  [    95% C.I.   ]      X
                                                                              
         =  .46990806
      y  = Pr(delays) (predict)
Marginal effects after logit

 

 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

36 
Published by SCHOLINK INC. 

Appendix 3 

                  Prob > chi2 =         0.0000
            Pearson chi2(146) =       333.19
 number of covariate patterns =       159
       number of observations =       832

Probit model for delays, goodness-of-fit test

 

                                                  
Correctly classified                        76.68%
                                                  
False - rate for classified -   Pr( D| -)   20.36%
False + rate for classified +   Pr(~D| +)   25.90%
False - rate for true D         Pr( -| D)   19.36%
False + rate for true ~D        Pr( +|~D)   27.12%
                                                  
Negative predictive value       Pr(~D| -)   79.64%
Positive predictive value       Pr( D| +)   74.10%
Specificity                     Pr( -|~D)   72.88%
Sensitivity                     Pr( +| D)   80.64%
                                                  
True D defined as delays != 0
Classified + if predicted Pr(D) >= .5

   Total           408           424           832
                                                  
     -              79           309           388
     +             329           115           444
                                                  
Classified           D            ~D         Total
                       True         

Probit model for delays

 

Appendix 4 

(*) dy/dx is for discrete change of dummy variable from 0 to 1
                                                                              
  famsup*   -.0207464      .04235   -0.49   0.624  -.103753   .06226   .653846
 loantyp*   -.3183615      .05019   -6.34   0.000  -.416741 -.219982   .673077
 postopp*    .0187582      .04288    0.44   0.662  -.065279  .102796   .653846
loanrenw*     .144945      .07151    2.03   0.043   .004796  .285094   .927885
 sponsor*    .5482105      .03145   17.43   0.000   .486577  .609844   .612981
relistat*    .3090923      .15512    1.99   0.046    .00506  .613125   .990385
maristat*   -.1054208      .06559   -1.61   0.108  -.233969  .023128    .90024
 shamexp*    .0157929      .04911    0.32   0.748  -.080466  .112052   .216346
indifail*    .1724996      .04951    3.48   0.000   .075457  .269542   .813702
jealneig*    .0290295      .07045    0.41   0.680  -.109049  .167108   .894231
 hapneig*    .1236404      .08974    1.38   0.168  -.052243  .299524   .956731
 endorse*     .055328      .05214    1.06   0.289  -.046866  .157522    .41226
                                                                              
variable        dy/dx    Std. Err.     z    P>|z|  [    95% C.I.   ]      X
                                                                              
         =  .46236814
      y  = Pr(delays) (predict)
Marginal effects after probit

 



www.scholink.org/ojs/index.php/ape               Advances in Politics and Economic                      Vol. 2, No. 1, 2019 

37 
Published by SCHOLINK INC. 

Appendix 5 

(*) dy/dx is for discrete change of dummy variable from 0 to 1
                                                                              
  famsup*    .0727744       .0206    3.53   0.000   .032404  .113145   .653846
 postopp*    .0027111      .02136    0.13   0.899  -.039148   .04457   .653846
maristat*     .098741       .0345    2.86   0.004   .031114  .166368    .90024
 shamexp*    .0879557      .02241    3.93   0.000    .04404  .131871   .216346
jealneig*   -.1363622      .03209   -4.25   0.000  -.199253 -.073472   .894231
 hapneig*   -.1597509      .03332   -4.79   0.000  -.225057 -.094444   .956731
 endorse*    .0098449      .01952    0.50   0.614   -.02842   .04811    .41226
                                                                              
variable        dy/dx    Std. Err.     z    P>|z|  [    95% C.I.   ]      X
                                                                              
         =  .51839399
      y  = Conditional mean of R (predict)

 

 

 

 


