Microfinance Interest Rate Puzzle: Price Rationing or Panic Pricing?

• Author: Lin Guo,Hoje Jo
• Published date: 01 April 2017
• Date: 01 April 2017
• DOI: http://doi.org/10.1111/ajfs.12167

Microfinance Interest Rate Puzzle: Price

Rationing or Panic Pricing?*

Lin Guo**

Korea University Business School, Republic of Korea

Hoje Jo

Leavey School of Business, Santa Clara University, United States

Received 28 June 2016; Accepted 20 December 2016

Abstract

We examine the relation between microfinance institution (MFI) interest rates, loan loss rates,

and regulations. Our results suggest that a positive and causal association exists where loan loss

rates drive MFI interest rates after controlling for confounding factors. This positive association

holds after accounting for potential endogeneity based on various econometric methods and

quasi-natural experiment. Although endogeneity cannot be ruled out fully, the resultsof our tests

suggest it plays a limited role. Our results support the price-rationing proposition that MFIs

ration credit and there is unmet demand at the market-clearing price. This positive relation is

more pronouncedin countries with individualisticand uncertainty avoiding cultures.

Keywords Microfinance; MFI interest rates; Loan loss rate; Regulation; MFI profitability

JEL Classification: G15, G18, G21, G38, I31, L26

1. Introduction

Microfinance is one of the few existing mechanisms that seeks to make credit available

to the most disadvantaged borrowers and entrepreneurs in the world (Garmaise and

Natividad, 2013). In short, microfinance aims to fight poverty by expanding access to

credit (Karlan and Zinman, 2008). Microfinance is essentially a recreated form of tra-

ditional finance in which one individual lends financial resources to another (partially

out of a sense of social responsibility to help a struggling entrepreneur better her own

life and partially out of a desire to make a small profit from idle funds). Microfinance

*This study was partially performed while Jo was visiting KUBS. Jo appreciates sabbatical

support from the Leavey School of Business at Santa Clara University and partial financial

assistance from the Asian Institute of Corporate Governance. The authors greatly appreciate

the valuable comments from an anonymous referee.

**Corresponding author: Lin Guo, Korea University Business School (KUBS), Korea Univer-

sity, 145, Anam-ro, Seongbuk-gu, Seoul, 02841, Korea. Tel: +82-2-3290-1380, Fax: +82-2-

3290-2552, email: gleileen@gmail.com.

Asia-Pacific Journal of Financial Studies (2017) 46, 185–220 doi:10.1111/ajfs.12167

©2017 Korean Securities Association 185

emerged as a solution to a lack of capital for those living in poverty in developing

economies but has since spread to developed economies where entrepreneurs find

micro loans difficult to secure (Freedman, 2000). Microfinance has also received its

share of criticisms and objections.

1

Although proponents of microfinance maintain that it has the power to trans-

form world poverty (Yunus, 2003, 2008), critics argue that microlenders charge

excessively high interest rates (Fernando, 2006; Hudon, 2009; Karlan and Zinman,

2009; MacFarquhar, 2010; Sun and Im, 2015) and that too much money is spent

chasing too few deals (Copeland, 2009; Wiesner and Quien, 2010), similar to the

venture capital and private equity industry (Gompers and Lerner, 2000). Research

finds that microfinance institutions (MFIs) typically have a much higher interest

rate policy than conventional commercial banks (Ashta, 2009; Armend

ariz and

Morduch, 2010), but there is no universally accepted rationale for this high rate.

Moreover, although microfinance’s popularity has grown, its provision has become

contested terrain (Khavul and Bruton, 2013; Khavul et al., 2013), as high interest

rates charged by MFIs increase the financial burden of entrepreneurs (Yunus, 2008;

Sun and Im, 2015). Thus, the trade-off between the interests of MFIs and their cli-

ents, or even the interests of other stakeholders, has been an issue for a long time

and may have led microfinance participants to become blas

e.

In this article, we first examine the empirical determinants of MFI interest rates

such as loan loss rates, governmental regulations, and other factors, as well as their

causal and endogenous impact on MFIs’ decisions regarding interest rates. We

examine these determinants to test whether the price-rationing effect of Stiglitz and

Weiss (1981, 1983) and possibly the adverse selection of Karlan and Zinman (2009)

or the panic pricing effect prompted by Karlan and Zinman’s (2009) moral hazard

problems is more representative of the microfinance industry. In addition, we

examine the interactive effect of governmental regulations and the loan loss rate on

MFI interest rates. Furthermore, we investigate the impact of country and cultural

factors on MFI interest rates. We maintain that the current understanding of hi gh

interest rates charged by MFIs is incomplete and that the rationale behind the high

rates remains puzzling. Although previous studies examine the empirical determi-

nants of MFI interest rates, none consistently controls for either endogeneity or

reverse causality (Cotler and Almazan, 2013; Roberts, 2013). Ignoring endogeneity

or reverse causality, however, could contaminate empirical inferences, leading to

biased conclusions. Previous studies also do not address the impact of cultural and

country factors on MFI interest rates. Our article examines the empirical

1

Microfinancing has existed in one form or another for centuries and in its simplest form

entails borrowing from friends or family on an interest-free basis. It gained popularity and

traction in the 1970s when Muhammad Yunus, who was later awarded the Nobel Peace Prize

for his efforts, developed the concept for the Grameen Bank in Bangladesh. Yunus’ mission

was simple: extend the ability of the poorest people to self-support through the use of small

and easy-to-understand loans (nobelprize.org).

L. Guo and H. Jo

186 ©2017 Korean Securities Association

determinants of MFI interest rates with an emphasis on the causal and endogenous

impact of loan loss rates and government regulations along with country and cul-

tural effects on MFI interest rates. Our article makes a contribution to the ongoing

debate by providing empirical evidence that offers insights into the question of MFI

interest rates and mitigates some of the measurement problems encountered by pre-

vious research due to endogeneity and reverse causality.

Because the interest rate charged to individual entrepreneurs or groups in

microfinance is difficult to observe in practice, we focus on the MFI nominal yield

on gross portfolio in percentage as well as on the real yield in percentage

2

as sec-

ond-best proxies of the MFI interest rate, following Sun and Im (2015). Using both

nominal and real interest rates as our empirical proxies, we examine the direction

and causality of the relation to determine whether the potential loan loss rate causes

MFIs’ high interest rates or the other way around.

Using data from the Microfinance Information Exchange (MIX) database and

employing various econometric methods, including ordinary least squares (OLS),

fixed-effect regressions, the Durbin–Wu–Hausman endogeneity test (see Greene,

2012), the dynamic panel system generalized method of moment (GMM), and

the panel vector autoregression (PVAR) model, we empirically find that although

MFI interest rates do not Granger-cause the MFI loan loss rate, the loan loss rate

is positively driving both nominal and real interest rates even after controlling

for other confounding factors. This new finding may reflect MFIs’ recognition

that the potential loan loss rate is one of the primary factors causing relatively

high interest rates, which supports the price-rationing effect (Stiglitz and Weiss,

1981, 1983) and the adverse selection of Karlan and Zinman (2009). In addition,

we find that the positive interest rate–loan loss rate association is more pro-

nounced in countries with relatively higher individualism, higher uncertainty

avoidance, and more human capital development economies. Furthermore, we

document that the positive effect of the loan loss rates on MFI interest rates is

mitigated if MFIs are required to register with their country’s government, report

financial performance to the government, or follow a particular set of regulations

or laws.

We review the related literature and develop hypotheses in the next section. The

following section discusses the sample, variable measurements, and research design.

Then, the subsequent sections present baseline empirical results, and a battery of

endogeneity tests and mitigation. We discuss the results and present conclusions in

the last section.

2

The nominal yield on gross portfolio in percentage represents the MFI interest rate and fees

on loan portfolio divided by gross loan portfolio, and the real yield on gross portfolio in per-

centage represents the nominal yield on gross portfolio minus inflation rate/(1 +inflation

rate).

Microfinance Interest Rate Puzzle

©2017 Korean Securities Association 187