DOI: 10.1111/1467-8268.

12446

ORIGINAL ARTICLE

Does financial development reduce the size of the informal

economy in sub‐Saharan African countries?

Henri Njangang | Luc Ndeffo Nembot | Joseph Pasky Ngameni

Faculty of Economics and Management,

LAREFA, University of Dschang,
Dschang, Cameroon
Correspondence
Henri Njangang, Faculty of Economics
and Management, LAREFA, University of
Dschang, Dschang, Cameroon.
Email: ndieupahenri@yahoo.fr
Abstract
This paper contributes to the literature in an attempt to understand the de-
terminants of the informal economy by investigating the relationship between
financial development and the size of the informal economy using an un-
balanced panel data of 41 sub‐Saharan African (SSA) countries over the period
1991–2015. The paper applies different techniques such as ordinary least
squares, fixed effects, system generalized method of moments, and quantile
regression. The results show that financial development and its square are
important determinants of the informal economy in SSA. Across a number of
specifications, we find that financial development reduces the size of the in-
formal economy and that there is a U‐shaped relationship between financial
development and the informal economy.

1 | INTRODUCTION

The informal sector is a major impediment to the ability of developing country governments to mobilize revenue and
gather the information needed to engage in effective development policies. The informal sector in sub‐Saharan Africa
(SSA) is one of the largest in the world with an average of 38% of GDP (Medina, Jonelis, & Cangul, 2017), employing no
less than 90% of nonagricultural employment, 66% of total employment, and accounting for 70% of production.
The proliferation of the informal sector, beyond its negative effects on public finances (Blackburn, Bose, &
Capasso, 2012) and the livelihood it provides to the poor and vulnerable, leads to congestion in cities, increases the
degradation of the environment (Biswas, Farzanegan, & Thum, 2012), increases debt (Elgin & Uras, 2013), promotes
corruption (Dreher & Schneider, 2010), reduces tourism (Badariah, Habibullah, & Baharom, 2015), increases in-
equalities (Rosser, Rosser, & Ahmed, 2000), maintains poverty (Kim, 2005), and establishes a state of insecurity in large
cities (Schneider, 2008). In view of its social, political, and economic consequences, it seems necessary for both
researchers and economic decision‐makers to find other determinants of the informal sector in order to set up effective
public policies to reduce its proliferation.
Over the last decades, there have been substantial studies on the determinants of the informal economy
(Dabla‐Norris, Gradstein, & Inchauste, 2008; Elbahnasawy, Ellis, & Adom, 2016; Medina et al., 2017). The tax burden is
considered in the literature as one of the most important causes leading to the proliferation of informality. Other factors
such as institutional quality and trade openness were identified as key determinants of the informal economy. In recent
years, limited access to credit has been identified as one of the major factors explaining the development of the informal
economy in Africa. Thus, facilitating access to small firms to credit can be an effective means to encourage informal
small firms to shift from informal economic activities to formal economic activities.
-------------------------------------------------------------------------------------------------------------------------------------------
© 2020 The Authors. African Development Review © 2020 African Development Bank

Afr Dev Rev. 2020;32:375–391. wileyonlinelibrary.com/journal/afdr | 375

376 | NJANGANG ET AL.

Although a large amount of literature has examined the impact of financial development along various dimensions
of economic development, namely: inequality (De Haan & Sturm, 2017), poverty reduction (Uddin, Shahbaz, Arouri, &
Teulon, 2014; Abosedra, Shahbaz, & Nawaz, 2016), productivity (Moretti, 2014), innovation (Ferreira, Manso, &
Silva, 2012), and most importantly economic growth (Ghirmay, 2004; Kouki, 2013; Walle, 2014), not too many studies
focus on the link between financial development and the informal economy. Theoretically, financial development
might affect the informal economy through several channels. For example, financial development, by lowering the
barriers to obtaining credit, facilitate entrepreneurs’ access to needed credit, increases the opportunity cost of producing
in the underground economy, and thus provides an incentive to informal firms to transition towards formal economy
(Blackburn et al., 2012; Bose, Capasso, & Andreas Wurm, 2012; Capasso & Jappelli, 2013).
There is little literature on the effect of financial development on the informal economy. Of the few studies on the
effect of financial development on the informal economy (see, e.g., Berdiev & Saunoris, 2016; Bose et al., 2012; Capasso &
Jappelli, 2013), none of them deal specifically with the case of SSA, which represents the area of the world where the
informal sector is more important and more damaging for public finance. Our goal is to bridge this gap by analyzing the
case of SSA countries. The contribution of this paper is therefore fourfold: first, this paper is one of the few attempts to
investigate the effect of financial development on the size of the informal economy in SSA countries. The few previous
studies have focused on developed or developing countries but have not specifically addressed the case of SSA countries.
To the best of our knowledge, this paper is the first to empirically estimate the effects of financial development on the
informal economy in SSA countries. Second, this paper uses the newly estimated data of the informal economy given by
Medina and Schneider (2017) that goes up to 2015, contrary to past studies. Previous studies used data from the informal
sector proposed by Elgin and Oztunali (2012), which only goes to 2009. Third, this paper goes beyond the linear
relationship that might exist between financial development and the informal sector and examines the nonlinear re-
lationship. Previous studies have only investigated the linear relationship between financial development and the informal
economy. Fourth, this paper is the first to combine parametric and nonparametric analysis to investigate the link between
financial development and the informal economy. Our baseline analysis is based on pooled OLS. However, it can be
argued that the OLS method imposes some restrictive assumptions on the model and does not account for endogeneity.
Next, to account for endogeneity we apply system GMM. Finally, we use the quantile regression (QR) to examine the
effect of the financial development on the informal economy at different intervals throughout the informal economy.
To sum up, the results show that financial development and its square are important determinants of the informal
economy in SSA. Across a number of specifications, we find that financial development reduces the size of the informal
economy and that there is a U‐shaped relationship between financial development and the informal economy.
The rest of this paper is organized as follows. Section 2 provides the theoretical framework and the empirical
literature. Section 3 describes the data and methodology. Section 4 presents and analyzes the results. Section 5 con-
cludes and makes some policy recommendations.

2 | THEORETICAL FRAM EWORK AND EMPIRICAL L ITERATURE

2.1 | Theoretical framework

Becker's (1968) work on the economics of crime provides the theoretical framework for analyzing the relationship
between financial development and the informal sector. The author shows that in the face of any illegal action, rational
individuals will make a comparison between the benefits of illegal activity and the costs of detection and punishment in
relation to the action. This analysis by Becker (1968) can be transposed in the case of the informal economy. Assuming
that one of the main causes of the informal economy is the payment of tax and regulation, it is clear that any rational
economic agent will make a comparison between the benefits of the informal (e.g. nonpayment of taxes) and costs
associated with the formal. Following the seminal work of Becker (1968), several authors have provided theoretical
explanations for the relationship between financial development and the informal sector. Among these authors we can
mention the works of Straub (2005), Blackburn et al. (2012), and Capasso and Jappelli (2013).
Straub (2005) develops a model in which he first presents the benefits and costs associated with both the formal and
informal economy. For Straub (2005), Adherence to costly registration procedures enables companies to benefit from
essential public goods, the enforcement of property rights, and contracts that make participation in the formal credit
market possible. In a framework of moral hazard with credit rationing, their decision is shaped by the interaction
between the cost of entering the formal system and the relative efficiency of formal and informal credit mechanisms
NJANGANG ET AL. | 377

and their related institutional arrangements. Blackburn et al. (2012) developed a simple model of tax evasion and
financial intermediation to explain the relationship between the informal economy and financial development. The
authors consider a situation where individuals have the choice between disclosing their entire wealth and benefiting
from the advantages offered by financial development in terms of credit, or concealing their wealth for the purpose of
tax evasion and losing the guarantee of obtaining a flexible condition loan. The authors go further and analyze how the
choice of such behavior can be influenced by a market situation which favors access to credit to finance formal
activities. They demonstrate in their analysis that financial development promotes greater disclosure of wealth and,
therefore, the lower the level of financial development, the greater the tax evasion, anything that promotes the informal
economy. The theoretical model developed by Capasso and Jappelli (2013) allows entrepreneurs to choose between low‐
efficiency technology and more advanced and rewarding technology. The choice of a high‐yield technology that
requires access to the credit market can be linked to the formal economy, while the choice of low‐return technology is
linked to the informal economy. Capasso and Jappelli (2013) conclude that financial development by reducing the cost
of credit increases the opportunity cost of informality, reduces credit rationing and incentives to operate informally, and
promotes investment in a high‐tech project.

2.2 | Brief literature review

Since the influential work of King and Levine (1993), several empirical and theoretical studies have analyzed the effect
of financial development on various dimensions of economic development, inter alia inequality, innovation, poverty
reduction and economic growth (El Menyari, 2019; Law, Kutan, & Naseem, 2018; Uddin et al., 2014). However, the
macroeconomic literature has paid little attention to the effect of financial development on the informal economy.
Among the studies that focus specifically on the impact of financial development on the informal economy, we
distinguish between macro and micro studies.
At the macro level, Bose et al. (2012) empirically investigate the dynamic relationship between banking develop-
ment and the shadow economy in 137 countries over the period 1995–2009. By using both cross‐sectional and panel
analysis, they found that both the depth and the efficiency of the banking sector matter in reducing the size of the
informal economy. Berdiev and Saunoris (2016), for a sample of 161 countries over the period 1960–2009, found that
financial development reduces the size of the shadow economy. In the case of Turkey, Bayar and Aytemiz (2017) show
that in the long run, financial development reduces the size of the informal economy and that the relationship is
unidirectional from financial development to the informal economy. Gobbi and Zizza (2012) investigate the relationship
between shadow economy and financial sector development in the Italian debt market and show that, while the
economy prevented the development of the financial sector, the effect of financial development on the shadow economy
is nonsignificant.
In line with macro‐level studies, the works done at the micro‐level confirm the importance of financial
development for the reduction of the informal economy. Beck and Hoseini (2014) use firm‐level survey data in
the Indian manufacturing sector and investigate the impact of financial deepening and bank outreach on the
informal economy. They found that bank outreach, by cutting barriers to entering the formal economy, has a
stronger effect on reducing the incidence of informality, especially for smaller firms. However, the authors show
that financial deepening has no significant impact on informal sector firms. Capasso and Jappelli (2013) use
microdata for Italy and analyze the relationship between financial development and the size of the underground
economy—they found that financial development can reduce tax evasion and the size of the underground
economy.

3 | METHODOLOGY

3.1 | Data

We investigate a panel of 41 SSA countries over the period 1991–2015 with data from different sources. The choice of
the selected countries and periodicity for this study are primarily dictated by the availability of reliable data over
constraints. The full description of the data is as follows:
378 | NJANGANG ET AL.

3.1.1 | Dependent variable

The dependent variable is the size of the informal economy (Informal 1). This variable is obtained from Medina and
Schneider (2017). These authors applied the Multiple Indicators Multiple Causes (MIMIC) modeling approach to
estimate the size of the informal economy as a percentage of GDP. This approach has two steps. In the first step, causes
and indicators of the informal economy are determined by considering the size of the informal economy as an
unobserved variable. In the second step, a structural model is used to estimate the coefficients of the causes and
indicators. For the robustness checks of our analyses, we use another measure of the informal economy (Informal 2)
coming from Elgin and Oztunali (2012).

3.1.2 | Independent variable

Our main independent variable is financial development. In this paper, we use two financial development indicators,
namely: domestic credit to the private sector as a percentage of GDP (credit) and Broad money as a percentage of GDP
(M2/GDP). These two financial development indicators are chosen according to financial literature on Africa, as the
financial sector in most African countries is dominated by the bank sector (Henri, Luc, & Larissa, 2019). These variables
are gathered from the World Bank's World Development Indicators (WDI). As a robustness check, this paper uses two
alternative measures of financial development from IMF, namely: the ratio of commercial bank assets divided by the
sum of commercial bank and central bank (Bank Assets) and the amount of liquid liabilities of the financial system as a
percentage of GDP. Figures 1 and 2 suggest a negative correlation between financial development indicators (M2/GDP
and credit) and the informal economy (Informal 1 and Informal 2).

3.1.3 | Control variables

Next to the financial development variables, we include nine control variables in order to ensure that estimated results
are not biased by omitted variables. They comprise: (a) GDP per capita growth rate (GDPPCgrowth); (b) human capital
(HK); (c) Population growth rate (PopGrowth); (d) Remittances; (e) Foreign direct investment; (f) Domestic investment
(DomInvest); (g) Foreign aid; (h) Savings; and (i) Inflation rate. Table A1 in the Appendix A provides a definition of the
variables and Table 1 presents the descriptive statistics of the variables used. Table 2 presents the correlation matrix.
70

70
60
60

50
50

40
40

30
30

20
20

0 50 100 150 0 50 100 150

M2/GDP Credit to private sector

Informal1 Fitted values Informal1 Fitted values

FIGURE 1 Financial development and the Informal 1 [Color figure can be viewed at wileyonlinelibrary.com]
NJANGANG ET AL. | 379

80

80
60
60

40
40

20
20

0 50 100 150 0 50 100 150

M2/GDP Credit to private sector

Informal2 Fitted values Informal2 Fitted values

FIGURE 2 Financial development and the Informal 2 [Color figure can be viewed at wileyonlinelibrary.com]

3.2 | Methodology

The purpose of this paper is to investigate the impact of financial development on the size of the informal economy in
41 SSA countries over the period 1991–2015. To this end, we estimate the following dynamic equation:

Informalit =β0 + β1 Informalit −1 + β2 FinDevit + β3 Xit + μi + vt + εit (1)

where Informalit is the size of the informal economy as a percentage of GDP, FinDevit is financial development, Xit
represents a vector of control variables, μi is an unobserved country‐specific effect, vt is time‐specific effect, and εit is the
error term.
To estimate our benchmark model (Equation 1), three different panel methods are used, namely, the ordinary least
squares (OLS), fixed effects (FE), and GMM. We first use the OLS estimator to estimate Equation (1). However, the OLS
model does not account for the country's FE and may suffer from omitted variables bias. To deal with country FE, we
subsequently applied FE model. However, when the FE technique is used to estimate this model, the estimated
coefficients are inconsistent and likely to be biased since the lagged value of the dependent variable is correlated with
the error term. In the presence of lagged dependent variable, Nickell (1981) argues that while OLS estimators may be
biased upwards, FE may be biased downwards. Moreover, given the existence of an endogenous relationship between
financial development and the informal economy, to estimate Equation (1) we apply the GMM proposed by Arellano
and Bond (1991), Arellano and Bover (1995), and Blundell and Bond (1998). In the literature, the GMM method is
frequently used to solve econometric problems such as heteroscedasticity, endogeneity, over‐identification, and validity.
Theoretically, the endogeneity issue may arise because of reverse causality, measurement errors, or omitted variable
bias. To better address the endogeneity issue, we estimate our model using a two‐step system GMM proposed by
Arellano and Bover (1995) and Blundell and Bond (1998).

4 | EMPIRICAL R ESULTS

This section presents the baseline as well as extended regression results of the effects of financial development on the
informal economy.
380 | NJANGANG ET AL.

TABLE 1 Summary statistics and list of countries

Panel A: Summary statistics

Variables Obs. Mean SD Min. Max.

Informal economy Informal 1 1,025 40.21 8.84 19.23 69.08
Informal 2 732 41.35 8.45 21.85 74.45
Financial development M2/GDP 973 28.91 22.07 2.19 151.55
Credit to private sector 971 17.73 22.69 0.41 160.12
Bank Assets 955 66.84 24.88 2.98 151.55
Liquid liabilities 955 28.68 21.35 3.29 137.73
Control variables GDP per capita growth 1,019 1.78 8.45 −47.81 140.50
Population growth 1,021 2.60 1.06 −6.18 7.92
Human capital 843 90.45 26.54 21.53 152.22
Remittances 715 4.17 8.76 0.00 71.80
Foreign direct investment 967 4.39 10.80 −82.89 161.82
For robustness check only Domestic investment 904 21.66 17.83 −2.42 219.07
Foreign aid 931 12.71 13.38 −0.25 181.19
Saving 904 9.00 23.87 −152.54 86.27
Inflation 952 85.07 1,128.09 −35.84 24,411.03
Panel B: List of countries (41)
Angola, Benin, Botswana, Burkina Faso, Burundi, Cabo Verde, Cameroon, Central African Republic, Chad, Comoros, Congo, Dem, Rep,
Congo, Rep, Côte d'Ivoire, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, The, Ghana, Guinea, Guinea‐Bissau, Kenya, Lesotho,
Liberia, Madagascar, Malawi, Mali, Mauritania, Mauritius, Mozambique, Niger, Nigeria, Rwanda, Senegal, Sierra Leone, South, Africa,
Tanzania, Togo, Uganda, Zambia, Zimbabwe.

4.1 | Baseline estimations

Table 3 reports the results of the preliminary estimation using OLS, FE, and GMM. We start with the simplest version of
the model by investigating if financial development reduces the size of the informal economy, ceteris paribus. Given the
debate and likely measurement errors in financial development, the estimation is carried out using two different
financial development variables, namely domestic credit to private sector, and M2/GDP.
The results in Table 3 clearly confirm our hypothesis: financial development reduces the size of the informal economy,
ceteris paribus. Both domestic credit and M2/GDP have negative coefficients and are highly significant in all cases. Results
suggest that countries with a more developed financial sector also have a smaller informal economy. For example, results in
Column (1) suggest that a 10‐unit increase in M2/GDP decreases the informal economy by 1.3 units and in Column (2), that
a 10‐unit increase in credit to the private sector decreases the informal economy by 1.62 units. This result can be explained
by the fact that financial development makes access to credit easier and cheaper, which facilitates entrepreneurship,
increases the opportunity cost of producing in the underground economy, and thus reduces the size of the informal
economy. Our results support the empirical findings of Bose et al. (2012), Capasso and Jappelli (2013), and Berdiev and
Saunoris (2016) which reveal a negative significant effect of financial development on the shadow economy.
However, the R2 of the regression is extremely low, meaning that a number of important variables are missing. The
absence of important determinants of the informal economy could bias the results in the case where the measurement
error is correlated with the omitted variables. In order to correct for this attenuation bias, we replicate regression by
using system GMM. The results presented in Columns (5) and (6) confirm the negative effect of financial development
on the size of the informal economy.
We additionally carry out the estimation with a number of control variables. As the results using FE and system
GMM were qualitatively similar, we report only the results for system GMM estimation in Table 4. We report in Table 4
a linear and nonlinear specification of the impact of financial development on the size of the informal economy. Results
from the linear specification clearly show that financial development variables (M2/GDP and credit) have the expected
negative coefficient and are highly significant, meaning that financial development reduces the size of the informal
economy. For example, results in Column (1) suggests that a 10‐unit increase in M2/GDP decreases the informal
 NJANGANG
ET AL.

TABLE 2 Correlation matrix

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15)
(1) Informal 1 1.000
(2) Informal 2 0.783 1.000
(3) M2/GDP −0.624 −0.593 1.000
(4) Credit −0.661 −0.509 0.808 1.000
(5) Bankasset −0.121 −0.263 0.435 0.363 1.000
(6) Liquidliabi −0.451 −0.554 0.803 0.488 0.104 1.000
(7) GDPPg −0.282 −0.107 0.217 0.161 0.108 −0.201 1.000
(8) HK −0.171 −0.245 0.194 0.135 −0.199 0.135 0.110 1.000
(9) PopG 0.642 0.390 −0.642 −0.610 0.075 0.093 −0.196 −0.314 1.000
(10) Remit 0.192 0.208 −0.007 −0.083 0.109 0.213 0.006 0.047 0.209 1.000
(11) FDI −0.023 −0.010 −0.012 −0.059 0.168 −0.195 0.152 −0.128 0.464 0.107 1.000
(12) Invest −0.240 −0.192 0.041 0.031 −0.043 0.031 0.089 0.335 −0.036 −0.115 0.254 1.000
(13) ODA 0.106 0.125 −0.267 −0.264 −0.024 −0.048 −0.029 0.033 −0.592 0.141 0.265 0.422 1.000
(14) Savings 0.018 0.091 0.093 0.146 −0.152 −0.339 −0.037 0.103 −0.029 −0.042 −0.342 0.267 −0.537 1.000
(15) Inflation 0.071 0.055 −0.078 −0.070 0.217 0.196 0.112 0.101 −0.551 0.629 −0.029 0.016 0.035 −0.148 1.000
|
381
382 | NJANGANG ET AL.

TABLE 3 Regression of financial development on the informal economy

Dependent variable: The size of the Informal 1

OLS Fixed effects System GMM

(1) (2) (3) (4) (5) (6)

M2/GDP −0.130*** −0.139*** −0.0162**
(0.0117) (0.0148) (0.00723)
Credit/GDP −0.162 ***
−0.222 ***
−0.0256***
(0.0109) (0.0177) (0.00506)
Lag informal 0.863*** 0.852***
(0.0222) (0.0228)
Constant 43.85*** 42.96*** 44.09*** 44.02*** 5.608*** 5.973***
(0.426) (0.314) (0.448) (0.339) (1.046) (0.996)
R2 0.113 0.186 0.086 0.145
Hansen test 0.223 0.176
AR(1) 0.00024 0.00026
AR(2) 0.743 0.744
Instrument 11 11
Observations 973 971 973 971 928 926

Note: The standard errors are reported in parentheses.

*Significance level at 10%.
**Significance level at 5%.
***Significance level at 1%.

economy by 1.19 units and in Column (2), that a 10‐unit increase in credit to private sector decreases the informal
economy by 0.825 units. Columns (1) and (2) present the nonlinear relationship between financial development and the
informal economy. The results suggest that the coefficients associated with the squared term of financial development
variables are negative and significant, meaning that financial development and the informal economy have a nonlinear
U‐shaped relationship. That is, financial development reduces the informal economy only up to a point; after this point,
it increases the size of the informal economy. In the literature, several authors (Law et al., 2018) have shown that there
is an inverted U‐shaped relationship between financial development and economic growth. These authors conclude
that financial development positively affects economic growth up to a certain threshold from which its effect becomes
detrimental to economic growth. The lack of economic growth has been identified as one of the causes of the informal
economy (Chen, 2012). Therefore, at a certain level of financial development, the decline in economic growth will result
in the rise of the informal economy.
The control variables have the expected signs. GDP per capita growth and human capital have the expected negative
sign suggesting that the higher growth rate of GDP per capita and quality human capital are associated with a lower size
of the informal economy. These findings confirm the idea that the lack of growth (Chen, 2012) and the higher cost of
credit (Capasso & Jappelli, 2013) increase the size of the informal economy. The coefficients associated with re-
mittances, population growth rate, and foreign direct investment have a positive sign and the sign is statistically
significant. The positive sign of remittances is consistent with the literature showing that remittances increase the size
of the informal economy. This result is explained by the fact that migrant remittances provide sufficient liquidity for
household consumption. In SSA countries, for the most part, goods are 70% produced by the informal sector. The
consequence is that the production of the informal sector is increasing. The result is consistent with Chatterjee and
Turnovsky (2018) and Njangang, Noubissi, and Nkengfack (2018), who show that remittances at both micro and macro
levels increase the size of the informal economy. The positive sign of the population is better understood, unlike that of
FDI. The increasingly growing population in SSA is seriously posing the problem of employment and thus un-
employment. Faced with the difficulty of African governments to provide employment for the growing number of
young people on the market, the informal sector thus remains the ultimate refuge of this idle workforce. The con-
sequence is an increase in the informal economy. One would have expected a negative effect of FDI on the informal
NJANGANG ET AL. | 383

TABLE 4 System GMM estimation with control variables

Dependent variable: Informal 1

(1) (2) (3) (4)

M2/GDP −0.119 ***
−0.0278 ***

(0.0255) (0.00874)
(M2/GDP)2 0.000851***
(0.000172)
Credit/GDP −0.0825*** −0.0560***
(0.0241) (0.0107)
***
(Credit/GDP)2 0.000504
(0.000178)
GDPPCgrowth −0.262*** −0.253*** −0.260*** −0.265***
(0.0445) (0.0388) (0.0463) (0.0286)
HK −0.0200 ***
−0.0235 **
−0.0229 ***
−0.0222**
(0.00533) (0.00906) (0.00482) (0.00855)
PopGrowth 0.248*** 0.237*** 0.291*** 0.192**
(0.0422) (0.0858) (0.0287) (0.0872)
** *** ***
Remittances 0.0199 0.0241 0.0247 0.0375*
(0.00864) (0.00698) (0.00822) (0.0220)
FDI 0.0487** 0.0634*** 0.0492** 0.0152
(0.0192) (0.0227) (0.0201) (0.0156)
*** *** ***
Lag Informal1 0.886 0.909 0.893 0.708***
(0.0343) (0.0264) (0.0333) (0.0549)
Constant 6.208*** 5.213*** 4.154* 17.06***
(2.122) (1.791) (2.283) (2.985)
Observations 560 560 560 560
Number of countries 40 40 40 40
AR(1) 0.0005 0.0004 0.0005 0.0001
AR(2) 0.964 0.809 0.980 0.999
Hansen OIR 0.666 0.522 0.624 0.173
Instruments 28 28 25 18
*** *** ***
Fisher 2,396 1,831 2,608 203.5***

Note: The standard errors are reported in parentheses.

*Significance level at 10%.
**Significance level at 5%.
***Significance level at 1%.

economy. However, this result can be explained by the structure of FDI destined for SSA countries. Most FDI is
concentrated in the primary sector, to the detriment of the manufacturing industry, which is a real job creator.

4.2 | Robustness checks

To investigate the sensitivity of the results reported in Tables 3 and 4, we conduct a number of robustness checks. For
this purpose, we evaluate our results successively, including additional control variables, using an alternative measure
of the informal economy, using system GMM with 5‐year average data, using alternative financial development
measures, and using a nonparametric analysis.
384 | NJANGANG ET AL.

TABLE 5 System GMM estimation with more control variables

Dependent variable: Informal 1

(1) (2) (3) (4)

M2/GDP −0.0664 ***
−0.109 ***

(0.00885) (0.0167)
Credit/GDP −0.0615*** −0.0855***
(0.00666) (0.0111)
GDPPCgrowth −0.237*** −0.240*** −0.237*** −0.219***
(0.0191) (0.0232) (0.0203) (0.0307)
HK −0.0302 **
−0.00312 −0.0313 **
−0.167***
(0.0135) (0.0101) (0.0144) (0.0198)
PopGrowth 0.243* 0.284* 0.0848 0.232
(0.143) (0.163) (0.121) (0.182)
** *** **
Remittances 0.0640 0.0461 0.0471 0.0430**
(0.0236) (0.0145) (0.0192) (0.0178)
FDI 0.0948*** 0.0303*** 0.0511*** 0.0613***
(0.0144) (0.0102) (0.0101) (0.0197)
Saving −0.0400 ***
−0.0299 ***

(0.00916) (0.00850)
ODA −0.0243*** −0.0237**
(0.00655) (0.0104)
DomInvest −0.0328 *
−0.0372
(0.0182) (0.0257)
Inflation −7.32e−05*** −7.12e−05**
(1.79e−05) (2.67e−05)
*** *** ***
Lag(Informal) 1 0.676 0.613 0.654 0.595***
(0.0251) (0.0405) (0.0264) (0.0510)
Constant 17.03*** 22.03*** 17.72*** 21.19***
(1.440) (2.304) (1.511) (2.718)
Observations 514 504 514 504
Number of countries 39 38 39 38
AR(1) 0.000421 0.000553 0.000501 0.000427
AR(2) 0.811 0.709 0.834 0.754
Hansen OIR 0.406 0.0426 0.469 0.106
Instruments 31 22 31 22
Fisher 240.3*** 326.8*** 454.4*** 324.7***

Note: The standard errors are reported in parentheses.

*Significance level at 10%.
**Significance level at 5%.
***Significance level at 1%.

4.2.1 | Including more control variables

The first set of robustness checks involves the inclusion of four more control variables, namely domestic investment, foreign
aid, savings, and inflation rate. The results reported in Table 5 are qualitatively similar to those reported in Table 4. More
specifically, all the coefficients associated with financial development variables (M2/GDP and credit) are negative and
NJANGANG ET AL. | 385

TABLE 6 Two‐step system GMM, using Informal 2

Dependent variable: Informal 2

(1) (2) (3) (4)

M2/GDP −0.0262 ***
−0.0109 ***

(0.00485) (0.00374)
Credit/GDP −0.00657*** −0.0208***
(0.00152) (0.00593)
GDPPCgrowth −0.0300** −0.0465*** −0.0241* −0.0381**
(0.0142) (0.0132) (0.0132) (0.0175)
HK −0.0251 ***
−0.0239 ***
−0.0283 ***
−0.0225***
(0.00461) (0.00596) (0.00399) (0.00547)
PopGrowth 0.248*** 0.237*** 0.291*** 0.192**
(0.0422) (0.0858) (0.0287) (0.0872)
*** ***
Remittances 0.0231 0.00853
(0.00642) (0.00260)
FDI 0.0555*** 0.0441***
(0.0118) (0.0123)
*** *** ***
Lag(Informal) 2 0.863 0.881 0.905 0.803***
(0.0159) (0.0194) (0.0148) (0.0171)
Constant 7.756*** 6.535*** 5.412*** 9.745***
(1.084) (0.981) (0.835) (1.100)
Observations 417 541 417 512
Number of countries 38 41 38 41
AR(1) 0.204 0.0515 0.322 0.0866
AR(2) 0.343 0.710 0.748 0.932
Hansen OIR 0.356 0.246 0.264 0.457
Instruments 22 22 25 21
*** *** ***
Fisher 1,320 1,544 6,709 736.5***

Note: The standard errors are reported in parentheses.

*Significance level at 10%.
**Significance level at 5%.
***Significance level at 1%.

statistically significant at the conventional levels, meaning that financial development reduces the size of the informal economy
in SSA countries. Therefore, the empirical results are robust to the inclusion of more control variables.

4.2.2 | Using an alternative measure of the informal economy

Second, we use an alternative measure of the informal economy proposed by Elgin and Oztunali (2012). Table 6 reports
the estimates of the system GMM with Informal 2 as the dependent variable. The number of sample countries is still the
same (41) for the analysis but the sample period for this new measure of the informal economy (Informal 2) only covers
from 1991 through 2008, the consequence is that the number of observations decreases in some specifications. The
results regarding the impact of financial development are consistent with the previous results. Both M2/GDP and credit
have negative and statistically significant coefficients, meaning that a more developed financial economy is associated
with the less informal economy.
386 | NJANGANG ET AL.

TABLE 7 Two‐step system GMM, 5‐year average data

Dependent variable: Informal 1

(1) (2) (3) (4)

M2/GDP −0.0949 ***
−0.0845 ***

(0.0106) (0.0102)
Credit/GDP −0.0391** −0.0245*
(0.0160) (0.0138)
GDPPCgrowth −0.196*** −0.239*** −0.251*** −0.114***
(0.0369) (0.0464) (0.0447) (0.0310)
HK −0.0798 ***
−0.0728 ***
−0.0958 ***
−0.0785***
(0.00964) (0.0131) (0.00914) (0.00903)
PopGrowth 1.428*** 0.235 2.438*** 0.178
(0.132) (0.486) (0.306) (0.390)
** ***
Remittances 0.0210 0.0922
(0.00794) (0.0169)
FDI 0.0449 0.0205
(0.0310) (0.0132)
*** *** ***
Lag(Informal) 2 0.854 0.984 0.558 0.828***
(0.0484) (0.0248) (0.0666) (0.0541)
Constant 8.037*** 4.733* 20.17*** 13.38***
(2.726) (2.394) (3.399) (3.219)
Observations 129 154 129 154
Number of countries 37 41 37 41
AR(1) 0.0345 0.0274 0.233 0.0236
AR(2) 0.0580 0.483 0.0315 0.503
Hansen OIR 0.580 0.686 0.738 0.810
Instruments 32 39 32 39
*** *** ***
Fisher 612.8 469.0 189.2 200.4***

Note: The standard errors are reported in parentheses.

*Significance level at 10%.
**Significance level at 5%.
***Significance level at 1%.

4.2.3 | Using system GMM with 5‐year average data

The third set of robustness checks involves the use of data over a 5‐year average instead of annual data. We divide the
sample period 1991–2015 into five nonoverlapping 5‐year periods to avoid the influence of idiosyncratic economic
dynamics at business cycle frequency, as well as to control for cyclical output movements. The empirical results
reported in Table 7 are quantitatively similar to those reported in Table 4. More specifically, all the coefficients
associated with financial development variables are still negative and statistically significant at the 1% level. For
example, the coefficient on M2/GDP in Column (1) is 0.0949; this implies that if M2/GDP increases by 1 unit, the size of
the informal economy will decrease by 0.0949 unit. On the other hand, the coefficient on credit, as shown in Column (3)
is 0.0391, meaning that a 1‐unit increase in credit will cause a 0.0391 unit decrease in the informal economy. Therefore,
the empirical results are robust to the alternative sample period. In terms of post estimation for all robustness checks,
the diagnostic tests show that all models are well specified. The difference‐in‐Hansen test is not rejected, and the
absence of second‐order autocorrelation AR(2) is found. Additionally, the number of instruments is less than the
number of countries, meaning that the problem of instrument proliferation does not exist.
NJANGANG ET AL. | 387

TABLE 8 System GMM with alternative financial development measures

Dependent variable: Informal 1

(1) (2) (3) (4)

Bank‐Assets −0.0209 ***
−0.0185 ***

(0.00618) (0.00612)
Liquid‐liabilities −0.0181*** −0.0561***
(0.00597) (0.0143)
GDPPCgrowth −0.223*** −0.277*** −0.241*** −0.230***
(0.0595) (0.0405) (0.0506) (0.0315)
HK 0.0187 0.0240 **
0.00485 −0.0410*
(0.0135) (0.0114) (0.0141) (0.0225)
PopGrowth −0.0542 −0.0936* 0.0418 −0.126
(0.0704) (0.0546) (0.0682) (0.526)
Remittances −0.0101 −0.0150** −0.00500 0.00512
(0.00702) (0.00557) (0.00621) (0.0161)
FDI 0.0463** 0.0139
(0.0213) (0.0131)
*** *** ***
Lag (Informal)1 0.899 0.926 0.890 0.728***
(0.0278) (0.0232) (0.0333) (0.0600)
Constant 3.989** 2.338 4.398* 16.68***
(1.883) (1.428) (2.372) (3.638)
Observations 560 560 567 567
Number of id 40 40 40 40
AR(1) 0.000415 0.000362 0.000564 0.000119
AR(2) 0.970 0.839 0.682 0.516
Hansen OIR 0.569 0.552 0.496 0.268
Instruments 22 25 22 18
*** *** ***
Fisher 1,916 1,655 842.0 84.29***

Note: The standard errors are reported in parentheses.

*Significance level at 10%.
**Significance level at 5%.
***Significance level at 1%.

4.2.4 | Using alternative measures of financial development

The fourth set of robustness is the use of two alternative measures of financial development obtained from the IMF. We
use the ratio of commercial bank assets divided by the sum of commercial bank and central bank (bank assets) and the
amount of liquid liabilities of the financial system as a percentage of GDP. The estimates results reported in Table 8
show that all the coefficients associated with financial development measures are negative and statistically significant at
the 1% level, supporting the previous established results in Table 4.

4.2.5 | Using a nonparametric analysis

The last set of robustness checks involves the use of a nonparametric analysis. The baseline results in Table 3 using
OLS estimator indicate that financial development is negatively correlated with the informal economy. Two
388 | NJANGANG ET AL.

TABLE 9 Quantile regression

Dependent variable: Informal 1

Q.25 Q.50 Q.75 Q.90

Panel A: M2/GDP
M2/GDP −0.199*** −0.212*** −0.191*** −0.191***
(0.0126) (0.0195) (0.0287) (0.0584)
GDPPCgrowth −0.235 ***
−0.233 ***
−0.202 *
−0.0938
(0.0502) (0.0776) (0.114) (0.232)
HK −0.0342*** −0.0314** −0.0167 −0.0200
(0.00803) (0.0124) (0.0182) (0.0371)
PopGrowth 0.665** 0.978** 1.330** 1.754
(0.276) (0.426) (0.627) (1.276)
Remittances −0.0388 −0.0296 −0.0851 −0.175
(0.0248) (0.0383) (0.0564) (0.115)
FDI 0.0313 0.0833 0.0320 −0.0410
(0.0339) (0.0524) (0.0771) (0.157)
Constant 42.41*** 44.98*** 47.09*** 53.87***
(1.269) (1.960) (2.882) (5.866)
Observations 606 606 606 606
Panel B: Credit/GDP
Credit/GDP −0.132*** −0.121*** −0.140*** −0.184***
(0.0126) (0.0137) (0.0197) (0.0386)
GDPPCgrowth −0.281*** −0.272*** −0.205* −0.136
(0.0670) (0.0728) (0.105) (0.205)
HK −0.0320*** −0.0241** −0.0196 −0.0238
(0.0107) (0.0117) (0.0167) (0.0328)
PopGrowth 1.704*** 1.211*** 1.209** 1.711
(0.362) (0.393) (0.565) (1.107)
Remittances −0.0229 −0.0819** −0.121** −0.203**
(0.0336) (0.0365) (0.0524) (0.103)
FDI 0.000541 −0.00137 −0.0330 −0.0665
(0.0449) (0.0489) (0.0702) (0.138)
Constant 36.16*** 40.47*** 45.06*** 52.84***
(1.563) (1.700) (2.441) (4.788)
Observations 605 605 605 605

Note: The standard errors are reported in parentheses.

*Significance level at 10%.
**Significance level at 5%.
***Significance level at 1%.

potential concerns with the results is unobserved individual heterogeneity and distributional heterogeneity, as
traditional OLS regression focus on the mean effects and does not take in account such heterogeneity. To control
for the distributional heterogeneity, the QR developed by Koenker and Bassett (1978) is used. By doing that, we
also test the robustness of the findings against the extreme value of our dependent variable. The results are
displayed in Table 9. Panel A displays the effect of M2/GDP on Informal 1, while Panel B presents the
effect of credit on Informal 1. The results are reported for the 25th, 50th, 75th, and 90th percentiles
of the conditional distribution. Overall, the empirical results indicate that the coefficients associated
with the various measures of financial development is negative and statistically significant, meaning that
financial development reduces the size of the informal economy. These results are consistent with the precedent
results.
NJANGANG ET AL. | 389

5 | CO NCL U S I O N A N D P O LI C Y I M P L I C AT I O N S

Does financial development reduce the size of the informal economy? To answer this important economic question, this
paper investigates the relationship between financial development and the size of the informal economy using an
unbalanced panel data set of 41 SSA countries over the period 1991–2015. Empirical evidence are based on OLS, FE,
system GMM, and the QR. The results suggest that financial development measured by broad money and domestic
credit to private sector has a highly statistically significant negative effect on the informal economy. The negative
impact of financial development on the informal economy is quantitatively important and robust to the inclusion of
more control variables, the use of another measure of informal economy, to the use of two alternative financial
development measures, to the use of an alternative year interval, and to the use of a nonparametric analysis. This
clearly suggests that a higher level of financial development reduces the size of the informal economy.
Based on the findings from our empirical analysis and given the main objective of this study, we can draw the
following policy implications. Governments willing to reduce the size of the informal economy should implement some
financial reform measures with the view of facilitating access to formal financing channels such as micro‐credit. In
addition, the governments of SSA countries must take other measures to control the use of credit granted to these
enterprises and assist them in the process of transition to the formal economy. The development of micro‐finance in
SSA was the solution found to facilitate the access of vulnerable groups to micro‐credit. However, the opportunistic
behavior of clients and the state's lack of guarantees pushed micro‐finance to tighten the conditions for access to credit
and consequently accentuated financial exclusion. African countries must therefore, in addition to favoring vulnerable
people's access to credit, ensure monitoring of the use of funds received to guarantee repayment at maturity.

ACKNOWLEDGMEN TS
The authors are indebted to the editor (John Anyanwu) and to two anonymous referees for their useful comments and
suggestions that substantially improved the quality of this paper. All remaining errors are ours.

ORCID
Joseph Pasky Ngameni http://orcid.org/0000-0003-0503-4878

REFERENCES
Abosedra, S., Shahbaz, M., & Nawaz, K. (2016). Modeling causality between financial deepening and poverty reduction in Egypt. Social
Indicators Research, 126(3), 955–969.
Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment
equations. The Review of Economic Studies, 58(2), 277–297.
Arellano, M., & Bover, O. (1995). Another look at the instrumental variable estimation of error‐components models. Journal of Econometrics,
68(1), 29–51.
Badariah, H. D., Habibullah, M. S., & Baharom, A. H. (2015). Does the shadow economy matter for tourism? International evidence. Taylor's
Business Review, 5(1), 1–9.
Bayar, Y., & Aytemiz, L. (2017). Financial development and shadow economy in Turkey. Unregistered Employment, 170, 1621–1632.
Beck, T., & Hoseini, M. (2014). Informality and access to finance: Evidence from India (CentER Discussion Paper Series No. 2014‐052).
Available at SSRN: https://ssrn.com/abstract=2491466 or http://dx.doi.org/10.2139/ssrn.2491466
Becker, G. S. (1968). Crime and punishment: An economic approach, The economic dimensions of crime (pp. 13–68). London: Palgrave
Macmillan.
Berdiev, A. N., & Saunoris, J. W. (2016). Financial development and the shadow economy: A panel VAR analyses. Economic Modelling, 57,
197–207.
Biswas, A. K., Farzanegan, M. R., & Thum, M. (2012). Pollution, shadow economy and corruption: Theory and evidence. Ecological
Economics, 75, 114–125.
Blackburn, K., Bose, N., & Capasso, S. (2012). Tax evasion, the underground economy and financial development. Journal of Economic
Behavior & Organization, 83(2), 243–253.
Blundell, R., & Bond, S. (1998). Initial conditions and moment restrictions in dynamic panel data models. Journal of Econometrics, 87(1),
115–143.
Bose, N., Capasso, S., & Andreas Wurm, M. (2012). The impact of banking development on the size of shadow economies. Journal of
Economic Studies, 39(6), 620–638.
Capasso, S., & Jappelli, T. (2013). Financial development and the underground economy. Journal of Development Economics, 101, 167–178.
Chatterjee, S., & Turnovsky, S. J. (2018). Remittances and the informal economy. Journal of Development Economics, 133, 66–83.
Chen, M.A. (2012). The informal economy: Definitions, theories and policies (Vol. 1, No. 26, pp. 90141‐4). WIEGO Working Paper.
390 | NJANGANG ET AL.

Dabla‐Norris, E., Gradstein, M., & Inchauste, G. (2008). What causes firms to hide output? The determinants of informality. Journal of
Development Economics, 85(1–2), 1–27.
De Haan, J., & Sturm, J. E. (2017). Finance and income inequality: A review and new evidence. European Journal of Political Economy, 50,
171–195.
Dreher, A., & Schneider, F. (2010). Corruption and the shadow economy: An empirical analysis. Public Choice, 144(1–2), 215–238.
El Menyari, Y. (2019). Financial development, foreign banks and economic growth in Africa. African Development Review, 31(2), 190–201.
Elbahnasawy, N. G., Ellis, M. A., & Adom, A. D. (2016). Political instability and the informal economy. World Development, 85, 31–42.
Elgin, C., & Oztunali, O. (2012), Shadow economies around the world: Model based estimates. Working Papers. Bogazici University
Department of Economics, 5(2012), 1‐48.
Elgin, C., & Uras, B. R. (2013). Public debt, sovereign default risk and shadow economy. Journal of Financial Stability, 9(4), 628–640.
Ferreira, D., Manso, G., & Silva, A. C. (2012). Incentives to innovate and the decision to go public or private. The Review of Financial Studies,
27(1), 256–300.
Ghirmay, T. (2004). Financial development and economic growth in sub‐Saharan African countries: Evidence from time series analysis.
African Development Review, 16(3), 415–432.
Gobbi, G., & Zizza, R. (2012). Does the underground economy hold back financial deepening? Evidence from the Italian credit market.
Economia Marche, Review of Regional Studies, 31, 1.
Henri, N., Luc, N. N., & Larissa, N., (2019). The long‐run and short‐run effects of foreign direct investment on financial development in
African countries. African Development Review, 31(2), 216–229.
Kim, B. Y. (2005). Poverty and informal economy participation: Evidence from Romania. Economics of Transition, 13(1), 163–185.
King, R. G., & Levine, R. (1993). Finance and growth: Schumpeter might be right. The Quarterly Journal of Economics, 108(3), 717–737.
Koenker, R., & Bassett, G., Jr. (1978). Regression quantiles. Econometrica: Journal of the Econometric Society, 46, 33–50.
Kouki, I. (2013). Financial development and economic growth in the North African region. African Development Review, 25(4), 551–562.
Law, S. H., Kutan, A. M., & Naseem, N. A. M. (2018). The role of institutions in finance curse: Evidence from international data. Journal of
Comparative Economics, 46(1), 174–191.
Medina, L., Jonelis, M. A. W., & Cangul, M. (2017). The informal economy in sub‐Saharan Africa: Size and determinants, Washington DC:
International Monetary Fund.
Medina, L., & Schneider, F. (2017). Shadow economies around the world: New results for 158 countries over 1991‐2015. CESifo Working Paper
No. 6430.
Moretti, L. (2014). Local financial development, socio‐institutional environment, and firm productivity: Evidence from Italy. European
Journal of Political Economy, 35, 38–51.
Nickell, S. (1981). Biases in dynamic models with fixed effects. Econometrica: Journal of the Econometric Society, 49, 1417–1426.
Njangang, H., Noubissi, E., & Nkengfack, H. (2018). Do remittances increase the size of the informal economy in sub‐Saharan African
countries? Economics Bulletin, 38(4), 1997–2007.
Rosser, J. B., Jr., Rosser, M. V., & Ahmed, E. (2000). Income inequality and the informal economy in transition economies. Journal of
Comparative Economics, 28(1), 156–171.
Schneider, F. (2008). Shadow economy, Readings in public choice and constitutional political economy (pp. 511–532). Boston, MA: Springer.
Straub, S. (2005). Informal sector: The credit market channel. Journal of Development Economics, 78(2), 299–321.
Uddin, G. S., Shahbaz, M., Arouri, M., & Teulon, F. (2014). Financial development and poverty reduction nexus: A cointegration and
causality analysis in Bangladesh. Economic Modelling, 36, 405–412.
Walle, Y. M. (2014). Revisiting the finance–growth nexus in sub‐Saharan Africa: Results from error correction‐based panel cointegration
tests. African Development Review, 26(2), 310–321.

How to cite this article: Njangang H, Ndeffo LN, Ngameni JP. Does financial development reduce the size
of the informal economy in sub‐Saharan African countries? Afr Dev Rev. 2020;32:375–391.
https://doi.org/10.1111/1467-8268.12446
NJANGANG ET AL. | 391

APPENDIX

T A B L E A1 Definitions of variables

Variables Definitions (measurement) Sources

Informal 1 Informal economy (% of GDP) Medina and Schneider (2017)
Informal 2 Informal economy (% of GDP) Elgin and Oztunali (2012)
(M2/GDP) Broad money (% of GDP) WDI
Credit/GDP Domestic credit to private sector (% of GDP) WDI
GDPPCgrowth GDP per capita growth (annual %) WDI
Human capital School enrollment, primary (% gross) WDI
Bank assets The ratio of commercial bank assets to the sum of commercial bank assets IMF
(% of GDP)
Liquid liabilities Liquid liabilities (% of GDP) IMF
Population Population growth (annual %) WDI
Remittances Personal remittances, received (% of GDP) WDI
FDI Foreign direct investment, net inflows (% of GDP) WDI
Inflation rate Inflation, consumer prices (annual %) WDI
Foreign aid Net ODA received (% of GNI) WDI
Domestic investment Gross fixed capital formation (% of GDP) WDI
Savings Gross domestic savings (% of GDP) WDI