Positive versus negative incentives for loan repayment in microfinance: A game theory approach

• Author: Julie De Pril,Marc Labie,Thomas Brihaye,Anaïs Périlleux
• Published date: 01 May 2019
• Date: 01 May 2019
• DOI: http://doi.org/10.1111/rode.12563

REGULAR ARTICLE

Positive versus negative incentives for loan

repayment in microfinance: A game theory

approach

Thomas Brihaye

1

|

Julie De Pril

2

|

Marc Labie

3

|

Anaïs Périlleux

4

1

Université de Mons (UMONS),

Complexys, Belgium

2

Université de Mons (UMONS), CERMi,

Complexys, Belgium

3

Université de Mons (UMONS), CERMi,

humanOrg, Belgium

4

Université Catholique de Louvain

(UCL), CIRTES, IRES, CERMi,

Belgium

Correspondence

Anaïs Périlleux, Université Catholique de

Louvain UCLouvain, CIRTES, IRES,

CERMi, Belgium

Email: anais.perilleux@uclouvain.be

Abstract

In most of the methodologies used so far by microfinance

institutions, negative incentives are predominant, which

can contrast with these institutions’social mission. This

paper investigates whether the microfinance industry

could benefit from using more positive incentives. The

main results of our game model are twofold. First, (posi-

tive or negative) incentives increase “on-time”repay-

ments. Second, the client is more likely to repay her loan

with the encouragement of a bonus scheme rather than

under the pressure of a sanction, regardless of the

amount. This paper therefore argues for the increased use

of positive incentives in the industry.

KEYWORDS

game theory, loan repayment, microfinance, positive incentives

1

|

INTRODUCTION

Four decades ago, economists started to confront the concept of the informal economy, which led

them to study informal markets, including informal financial markets. Looking at older practices

inherited from informal markets and from long-standing institutions, some actors designed new

ways of attending to the financial needs of excluded populations through what are now known as

microfinance institutions (MFIs). A huge variety of schemes and institutions were established

around the world, with village banking, cooperatives, non-governmental organizations (NGOs),

non-banking financial institutions, and even made-up banks being created to deliver microfinance

services.

From these multiple innovations, some standard practices progressively emerged in microfi-

nance to become what is called “mainstream microfinance”. Mainstream microfinance is most of

DOI: 10.1111/rode.12563

Rev Dev Econ. 2019;23:577–597. wileyonlinelibrary.com/journal/rode © 2018 John Wiley & Sons Ltd

|

577

all a question of methodologies and product designs applied in order to deliver financial services

to financially excluded people by overcoming transaction costs and the information asymmetry

issues that prevent banks from serving those people. It is made up of various common procedures

and incentive mechanisms that ensure the right selection and assistance of microfinance clients to

push them to respect their commitments, such as reimbursing their loans. By applying these proce-

dures and incentive mechanisms, MFIs now serve over 211 million people around the world (ac-

cording to the Global Microcredit Summit in 2015). At the same time, credit constraints remain a

major issue in developing countries (Armendáriz & Labie, 2011). The number of people who

could benefit from greater financial inclusion is estimated at between 500 million and 2.5 billion,

so there is still a long way to go.

There is a predominance of “negative incentives”in MFIs’methodologies. As shown by Galari-

otis, Villa, and Yusupov (2011) in their critical literature survey of microlending contracts, in the

face of pressure to achieve sustainability, MFIs adopt anti-social enforcement methods, which are

in contradiction with their social mission. These authors call for more research into ways to over-

come this paradox.

In line with this concern, this paper discusses the extent to which the increased use of “positive

incentives”may actually help to further stimulate more mainstream microfinance. Cadena and

Schoar (2011) evaluated the effectiveness of three positive incentives to encourage timely repay-

ments in Uganda. They found in particular that, compared with the control group, customers facing

positive incentives were 8 percent more likely to repay all their installments on time. Moreover, on

average, their arrears per month were 2 days shorter. Meanwhile, Peysakhovich (2014) studied dif-

ferent kinds of commitment contracts. He compared ‘carrot’contracts (rewards for good behavior)

with ‘stick’contracts (fines for bad behavior) and showed that, from a welfare point of view, deci-

sion-makers have a strict preference for the carrot rather than the stick.

Incentive mechanisms in microfinance have been widely analyzed and modeled. But to the best

of our knowledge, no paper pays specific attention to the positive versus the negative aspect of

incentive mechanisms on borrowers’repayments. In this paper we use game theory to investigate

and model this issue. We consider two ex-post incentive mechanisms: a positive incentive (where

the MFI grants a bonus for on-time repayments) and a negative incentive (where the MFI imposes

a sanction for late repayments). We model a commercially oriented MFI, which tries to cover its

costs, as is now largely the case in microfinance. This MFI provides individual loans. As exempli-

fied by Grameen II, in recent years, individual lending has dramatically increased in microfinance,

having now overtaken group lending. MFI clients generally prefer to borrow individually rathe r

than in groups (Armendáriz & Morduch, 2000). MixMarket data on 1,031 of the largest MFIs

show that individual lending represents 79 percent of the gross loan portfolio of those institutions.

1

Our results appear to be consistent with previous empirical findings, with our model showing that

positive incentives are a more effective way to encourage repayments than negative incentives.

This is of importance for product design and public policies: more attention should be given to

positive incentives and to the conditions in which they are provided.

The rest of this paper is organized as follows. Section 2 provides an overview of the mecha-

nisms used by mainstream microfinance to serve unbankable people and the predominance of neg-

ative incentives. Section 3 discusses models of microfinance incentives and the need to consider

negative and positive incentives as two different types of mechanism. Section 4 presents the model

comparing the impact of positive and negative incentives on loan repayment using a two-player

game. Section 5 presents a numerical example in order to give a concrete idea of how costly it is

for the MFI to implement a bonus scheme. Section 6 sums up the results obtained from the model

and provides some concluding remarks.

578

|

BRIHAYE ET AL.