1

NON-BANK FINANCIAL INSTITUTIONS AND ECONOMIC GROWTH:

EVIDENCE FROM SELECTED AFRICAN COUNTRIES

By

Ronald Rateiwa*

University of Stellenbosch Business School

PO Box 610, Bellville 7530, South Africa

Tel: (27) 745853467

rrateiwa@hotmail.com

and

Meshach Jesse Aziakpono University of Stellenbosch Business School

PO Box 610, Bellville 7530, South Africa

Tel: (27) 21 918 4261

meshach.aziakpono@usb.ac.za

Abstract

This paper uses Johansen cointegration and the vector error correction model within a country

specific setting to empirically test the existence of a long-run equilibrium relationship between

economic growth and non-bank financial institutions (NBFIs) and the causality thereof. The

empirical assessment is based on evidence from selected African countries over the period

1971-2013. The results show that a strong long-run relationship between NBFIs and economic

growth exist in Egypt and South Africa. Evidence in respect of Nigeria is weak. Thus, the study

reveals that, in countries with more developed financial systems, the role and importance of

NBFIs to the economic growth process is more pronounced.

JEL Classification: E2, G1, G2, O4

Key Words: Non-bank Financial Institutions, economic growth, financial development, Egypt, Nigeria,

South Africa

Corresponding author

September 2015

2

1. INTRODUCTION

As the 2015 deadline for the Millennium Development Goals (MDGs) approaches, a new

development agenda to build on the achievements of the MDGs and address new challenges in

the post 2015 era is being finalised (World Bank, 2013). This post-2015 transformative

development agenda – termed the Sustainable Development Goals (SDGs) - requires that

available resources be used more effectively and that additional financing be mobilised from

the private sector. However, given the fragility of financial markets in the wake of the global

crisis, the availability of long-term financing required to support productive investment has

been constrained. Specifically, traditional bank lending has slowed down as banks recover from

the financial crisis and adjust to tighter regulatory requirements - mostly emanating from the

Basel III requirements. To this end, the development of non-bank financing has become

imperative (World Bank, 2013). This paper explores the role of non-bank financial institutions

(NBFIs)as a source of long term financing by providing an empirical assessment of the linkage

between NBFIs and economic growth (the finance-growth nexus) using evidence from selected

African countries (Egypt, Nigeria and South Africa) over the period 1971 to 2013.

NBFIs are financial institutions that do not have a full banking licence and thus cannot take

deposits. However, they both compete and complement banking institutions by providing

alternative financial services such as contractual savings (pension funds and insurance

companies), investment intermediaries (finance companies, mutual funds and money market)

and consumer credit (Mishkin, 2007 and World Bank, 2015).

A cross-reading of literature revealed that, although the importance of finance for economic

growth has been theoretically postulated and empirically supported, the results are not

conclusive (Gertler, 1988; Levine, 1997 and 2004; Aziakpono, 2011; Harris, 2012 and Levine

et al, 2013). Two primary reasons for this emerge. Firstly, using different indicators of financial

development produces different results. Levine et al (2013) rightly argued that financial

markets are multi-dimensional and each measure captures a unique separate facet of financial

systems. The literature shows that in most studies, NBFI is not included as part of indicators

of financial development, which exclusion implies that the effect of financial development of

economic growth is underestimated.

Secondly, the results from panel/cross country studies should be interpreted with caution

(Eschenbach, 2004; and Ang, 2008). Cross-country studies hide cross-country differences

(Arestis et al, 2010; Ahmed and Wahid, 2010).

However, given the emergence of NBFIs as an important source of long-term productive

capital, in the face of constrained traditional bank financing, it is imperative that the finance-

growth nexus be re-assessed using NBFIs indicators. Furthermore, a cross-reading of literature

revealed that none of the empirical studies specifically investigating the relationship between

economic growth and NBFIs reported on causality and the direction thereof, between these two

variables. This paper is a modest effort in this regard. It will use NBFI as an indicator of

3

financial development to re-assess the finance growth nexus using time-series data from

selected African countries (Egypt, Nigeria and South Africa). In addition, the paper makes

another contribution to finance-growth literature by testing causality between economic growth

and NBFIs. In carrying out this assessment, we will use cointegration and error correction

modelling techniques within a country-specific setting.

Over the recent past, assets under the management of NBFIs significantly grew. Specifically,

data from the Global Shadow Banking Monitoring Report 2014shows assets of (i) insurance

companies increased from US$13.7 trillion in 2002 to more than US$27 trillion in 2013; (ii)

pension funds increased from US$11.7 trillion in 2002 to more than US$27 trillion in 2013

while (iii) those of other NBFIs increased from US$26.4 trillion to more than US$75 trillion

over the same period (see Annexure 1). Other NBFIs include money market funds, finance

companies and structured finance vehicles. In addition, NBFI assets constitute 25% of total

financial system and more than half of the banking institutions’ assets (Financial Stability

Board, 2014).

The emergence of NBFI in Africa has also not been less evident. Available data also shows

that, although the share of NBFIs as a percentage of GDP for sub-Sahara African countries has

been predominantly smaller than that of high income countries and the world average; after the

financial crisis it increased to above that for high income countries and the world average (see

Annexure 2). This is indicative of their potential to as a pool of non-bank long term finance to

invest in productive capital, especially in a period of constrained bank lending. The natural

question that follows is why focus the study on these three countries.

Nigeria, South Africa and Egypt are the three biggest economies in Africa based on World

Bank GDP ranking data (2015).1 They constitute almost half (49%) of Africa’s sum total GDP.

Secondly, to show the importance of these countries to financial systems in Africa, research

shows that South Africa, Egypt and Nigeria have the biggest share of banking assets in Africa

(KPMG Africa Limited, 2013). More specifically, collectively they constitute 73% of banking

assets in Africa.2 Thirdly, these three countries make up approximately 60% of all

infrastructure investments in Africa, which investments are predominantly funded through

NBFIs (Mo Ibrahim Foundation, 2013).3 Fourthly, a recent study by the United Nations

Economic Commission for Africa (UNECA) (2015) showed that Nigeria, Egypt and South

Africa constitute more than 56% of illicit financial flows4 in Africa.5 Lastly, these countries

have the oldest capital markets in Africa, and thus it implies that they have data available over

a period longer than any other country in Africa.

1 Accessed at http://data.worldbank.org/data-catalog/GDP-ranking-table on 25 August 2015 2 South Africa accounts for 51%, Egypt, 13% and Nigeria 9% of the total bank assets of Top 100 Banks in Africa.

(Financial Services in Africa Report by KPMG Africa Limited, 2013: 7) 3 “Africa Ahead: The Next Fifty Years” report 4 Illicit financial flows emanate from money laundering, corruption, tax abuse and trade mispricing. The estimate

of illicit financial flows from Africa over the period 1970-2008 is three times Africa’s current external debt 5 “Illicit Financial Flows: Report of the High Level Panel on Illicit Financial Flows in Africa”

4

The remainder of the paper is organised as follows: Section 2 will provide the theoretical link

between finance and economic growth and the role of NBFIs in facilitating economic growth;

Section 3 will provide the theoretical framework linking NBFIs to economic growth and a

context of the development of NBFIs in Egypt, Nigeria and South Africa; Section 4 presents

the methodology; Section 5 presents the empirical results; and (v) Section 6 concludes.

2. FINANCIAL SYSTEMS AND ECONOMIC GROWTH

This section provides a brief genesis of the theoretical framework underlying the finance-

growth nexus and a review of empirical work thereto. The section will conclude by reviewing

the importance of NBFIs to the finance-growth nexus.

2.1. Genesis of the finance-growth debate

Debate on the finance-growth nexus is more than a century old. Early proponents of the role of

finance in economic growth process include Bagehot (1874) and Schumpeter (1912). Bagehot

argued that finance was required to finance invention and development of the railway system

and also to facilitate trade in England. However, he caution that a financial system is a

combination of ‘economical power and economical delicacies’; that is, it has power to facilitate

trade, yet prone to collapse if not managed properly (Bagehot, 1874: 4). In support of Bagehot’s

views, Schumpeter also argued that finance is ‘nothing but a means of diverting the factors of

production to new uses, or of dictating a new direction to production’ (Schumpeter, 1912: 161).

Another early researcher on the relationship between finance and economic growth was Fischer

(1933). In his paper, Fischer argued that in the absence of the over-indebtedness which was

caused by the financial sector; the depth and duration of the economic crisis should not have

been as severe as seen in 1837, 1873, or 1929-33. Thus, he further explained that the economic

crisis was caused by over-investment and over-speculation done using borrowed money

obtained from the financial system.

In the 1950s, the finance-growth debate was revived by Gurley and Shaw (1955), among others,

who argued that without financial intermediation, economic development is retarded. Financial

intermediaries quicken the growth in debt, relative to income, thus creating more loanable

funds. On this basis, they concluded that debt accumulation - which is facilitated by banks - is

part of the growth process. Gurley and Shaw was followed by Goldsmith (1959), who also

supported the importance of finance to the economic growth process when he concluded that

the structure of the financial system will influence the amount and character of funds available

to finance the capital formation process.

Interest in the finance-growth debate continued into the 1970s and 1980s, where the financial

repression concept was popularised by Mackinnon (1973) and Shaw (1973). The two

researchers concurred that financial repression was a demonstration of economic myopia and

political weakness on part of the incumbent government. They argued that repression was seen

as the main cause for the lack of economic growth in less developed countries. Thus, they

5

concluded that financial markets should be liberalised and be allowed to reflect their true prices

which are relevant to the economy.

Although Mackinnon and Shaw’s work was criticised for lack of statistical evidence, their work

remains relevant to this day. Specifically, Stiglitz (2000) argued that financial liberalisation in

an economy where information asymmetry and poor corporate governance are endemic, will

not be welfare enhancing. This may explain the reason why most African countries did not

realise economic benefits after liberalising their economies in the 1980 under the economic

restructuring programmes. Secondly, Stiglitz (1994) also argues that financial regression can

be welfare enhancing, if it reduces the cost of capital, which capital will be used to fund

productive assets. A classic example is China, who managed to achieve tremendous economic

growth under repressive financial regimes.

The following section provides a discussion of the theoretical model of the finance-growth

nexus and selected review of empirical work on the debate.

2.2. Theoretical model underlying the finance-growth nexus

One of the earliest theoretical frameworks linking financial development to economic growth

was developed by Pagano (1993).In his paper, Pagano used the endogenous growth model (AK

model) to show that improvements in financial intermediation will increase the amount of

capital allocated to investment, the marginal productivity of capital and also the savings rate.

In this way, Pagano argued that financial systems will influence the growth rate of an economy.

Pagano’s work was immediately followed by the work of King and Levine (1993a and b),

whose seminal work inspired further empirical work on the finance growth-nexus (Aziakpono,

2011 and Harris, 2012). The emerging consensus from these studies, among others, is that the

financial system performs a number of functions, which functions lead to increased productive

capacity of the economy thus increasing economic growth (see Figure 1 below). The schematic

Figure 1 shows that market frictions create the need for developing financial intermediaries. In

order to overcome such frictions, financial intermediaries carry out a number of functions such

as mobilisation of savings and risk management. By carrying out these functions, financial

intermediaries improve the rates of capital accumulation, technological innovation and total

factor productivity; thus influencing economic growth. We illustrate this process in Figure 1

below.

6

Figure 1: Functions of financial markets and intermediaries

Source: Based on Levine 1997

3. NBFIs AND ECONOMIC GROWTH

The traditional finance-growth debate focused on the role of banks and stock markets in

fostering economic growth. Specifically, only bank and stock market development indicators

are used in the arguments. However, given the emergence of NBFIs as an alternative source of

capital to increase the productive capacity of an economy, it is imperative that effort be

committed to understand more about the role of NBFIs in the finance-growth debate.

As indicated above, NBFIs are financial institutions that do not have a full banking licence and

thus cannot take deposits. However, they both compete and complement banking institutions

by providing alternative financial services. Mishkin (2007) classifies NBFIs into two main

categories, namely contractual savings institutions and investment intermediaries.

Contractual savings institutions are financial intermediaries that obtain funds from individuals

and institutions on a contractual basis, at regular intervals. They mostly invest in corporate

bonds, stocks and mortgages. These institutions include life insurance companies, short term

insurance and pension funds. On the other hand, investment intermediaries are financial

institutions that facilitate purchase of capital and money market instruments. These include

finance companies, mutual funds and money market institutions. Table 1 below shows the

characteristics of these institutions in terms of the assets and liabilities they hold.

Table 1: Characteristics of NBFIs

Type of intermediary Liabilities (Source of

Funds)

Assets (Use of Funds)

Contractual savings

institutions

Life insurance companies Premiums for policies Corporate bonds and

mortgages

7

Short-term insurance

companies

Premiums for policies Corporate bonds, stocks

and government bonds

Pension funds Employer and employee

contributions

Corporate bonds and

stocks

Investment

intermediaries

Finance companies Commercial paper, stocks

and bonds

Consumer and business

loans

Mutual funds Shares Stocks and bonds

Money market Shares Money market

instruments

Source: Mishkin (2007: 40)

By intermediating funds between surplus and deficit units, NBFIs facilitate efficient allocation

of capital, which leads to higher production and growth of the economy.

Literature reviewed in this paper show that the linkage between NBFIs and economic growth

can be both direct and indirect (Holzmann, 1997; Vittas, 1997; Davies and Hu, 2008; Haiss

and Sumegi, 2008; Meng and Pfau, 2010; Aldermen and Yemtsov, 2013; Nassr and Wahinger,

2014; Liang and Reichert, 2015; Sufian and Majid, 2015). Direct in the sense that NBFIs can

directly influence savings, investment, risk allocation and total factor productivity, thus

enhancing economic growth. On the other hand, the linkage can be indirect through their

influence on the development of the bank and capital (stock and bond) markets, which in turn

influence economic growth. We present this argument schematically in Figure 2 below.

Figure 2: The linkage between NBFIs and economic growth

Source: Author’s analysis

Meng and Pfau (2010) argued that NBFIs can influence economic growth indirectly through

their impact on other institutions of the financial markets. One way through which this will

8

happen is by increasing competition to traditional banks in the loanable funds markets.

Increased competition will force banks to increase the volume of loans by lowering interest

rates in order to maintain their market shares (Sufian and Majid, 2015). In addition to increased

competition, the presence of NBFIs will increase the absolute volume of loanable funds in the

financial markets thus improving liquidity of the financial sector (Haiss and Sumegi, 2008).

However, it also has to be noted that if the growth in volume of loanable funds is too rapid and

is not properly regulated and monitored, it may create conditions susceptible to financial crisis.

Liang and Reichert (2015) warned that if NBFIs are not properly regulated they allow excessive

risk appetite which may have disastrous consequences for both the financial sector and the

economy.

The other way in which NBFIs can indirectly influence economic growth indirectly is through

the capital markets. Although the allocation of assets to different instruments in certain markets

varies from country to country depending on history and regulation, the argument is that NBFIs

can enhance depth, liquidity and efficiency of the capital markets. Specifically, NBFIs have

the potential to catalyse modernisation of capital markets, efficient trading and settlement

systems. Furthermore, institutional investors can also help improve governance and

information disclosure as they are becoming more actively involved in monitoring performance

of their investments (Vittas, 1997).

The direct linkage between NBFIs and economic growth occurs when NBFIs facilitate

mobilisation of savings, access to finance by small to medium enterprises (SMEs), capital

accumulation, risk management and also provide an incentive for investing in long-term

productive capital (Haiss and Sumegi, 2008; Aldermen and Yemtsov, 2013). By improving

access to finance by SMEs, NBFIs will help SMEs build resilience during economic shocks,

thus enhancing their contribution to employment and poverty reduction. NBFIs also provide

different insurance instruments which may provide incentives for individuals and corporates to

invest in long-term assets. In addition, the long-term nature of liabilities for NBFIs also

facilitates longer investment horizon. Longer investment horizon is suitable for productive

assets such as infrastructure, which enhance economic growth.

The following section will review selected empirical evidence on the linkage between NBFIs

and economic growth.

3.1. Empirical evidence on NBFIs and economic growth

As indicated above, there are few empirical studies that focused on the relationship between

NBFIs and economic growth. However, like other empirical literature on the finance-growth

nexus, results on the linkage between NBFIs and economic growth are mixed. In addition, they

also reveal a weak association between NBFIs and economic growth.

One of the early empirical studies that examine the linkage between NBFIs and economic

growth focused on Chile, following the implementation of its popular pension reform

programme. The study showed that, after the pension reforms in 1981, savings generated by

the NBFIs were negative until around 1988. This suggests that the influence of NBFIs on

9

economic growth will only be significant if the development of NBFIs has achieved a certain

threshold (Holzmann, 1997). These findings were also echoed by a more recent study on

Malaysian NBFIs which found that although the overall effect of NBFIs was negative,

evidence showed that NBFIs that are relatively smaller have limited capabilities to adopt

advanced technologies which would push their productivity frontier (Sufian and Majid, 2015).

In this way, their capability to influence economic growth is limited.

To further demonstrate the importance of size on the ability of NBFIs to influence economic

growth, a study of the NBFIs in the United State of America (US) revealed that as the NBFIs

grew in sheer size, their ability to influence capital markets also increase (Vittas, 1997).

Specifically, the study showed that because of their growing size, pension funds in the US have

undermined the dominance of capital markets by a few players. In addition, they also

contributed to the advent of competitive bidding for corporate issues and restructuring of stock

exchanges which included the abolition of minimum commissions for equity trading. These

findings are also in line with those from a more recent study on 29 European countries (Haiss

and Sumegi, 2008). The study on European NBFIs found evidence of significant positive

association between NBFIs and economic growth, only for 15 countries, which countries had

more developed financial systems.

In China and India, empirical studies show that access to insurance has a positive effect on the

level of production and risk appetite (Aldermen and Yemtsov, 2013). Specifically, the study

records that evidence from a randomised experiment in India showed that farmers with rainfall

insurance tend to switch to more profitable but riskier crops. In China, when farmers accessed

insurance, they raised pig production. However, other studies on the same question but using

different indicators (assets and liabilities) of NBFIs in China and Malaysia did not find

evidence of a positive association between NBFIs (Cheng and Degryse, 2010 and Sufian and

Majid, 2015). Cheng and Degryse found that growth of NBFIs in China was not correlated with

economic growth. They argue that this is due to the fact that NBFIs in China are less developed

and they have not benefited from the deregulation and liberalisation processes, which other

financial institutions such as banks benefited from. In Malaysia, Sufian and Majid found that

NBFIs experienced a productivity regress, which would adversely affect economic growth.

Lastly, an empirical cross-country study which included Egypt, Nigeria and South Africa found

a negative relationship between NBFIs and economic growth for both developed and emerging

market countries (Liang and Reichert, 2015). The argument for their finding was that that

NBFIs are not properly regulated hence they allow excessive risk appetite which may have

disastrous consequences for both the financial sector and the economy, especially in the wake

of the recent global financial and economic crisis. In addition, two separate country specific

studies found that there is no long-run relationship between economic growth and NBFIs when

using assets and liabilities of finance and discount houses as indicators of NBFIs (Osuala and

Odunze, 2014 and Ndugbu at al, 2015). However, their study found a positive relationship

between insurance and economic growth, suggesting that only certain aspects of NBFIs in

Africa are positively associated economic growth.

10

A cross-reading of literature revealed that none of the empirical studies specifically

investigating the relationship between economic growth and NBFIs reported on causality and

the direction thereof, between these two variables. As indicated above, this paper makes

contribution to finance-growth literature by testing causality between economic growth and

NBFIs. In the following section, we review the development of NBFIs in Egypt, Nigeria and

South Africa.

3.2. Development of NBFIs in Africa

As highlighted above, the importance of NBFI in Africa has been steadily increasing, pointing

to an alternative source of long-term financing. Figure 3 shows growth in the assets of NBFIs

expressed as a percentage of GDP for Egypt (NBF_E), Nigeria (NBFI_N) and South Africa

(NBFI_SA). The values are shown on the left-hand axis for Egypt and Nigeria, while that for

South Africa is displayed on the right-hand axis.

Figure 3: Assets of NBFI expressed as percentage of GDP for Egypt, Nigeria and South

Africa: 1971-2013

Source: Global Development Finance Database (2013), Central Banks of Nigeria, Egypt and South Africa

Figure 3 above shows that, although NBFI assets for Egypt appear to be fluctuating over time,

they have maintained an upward trend; recovering from a low of under 2% in 1979 to more

than 7% in 2012. NBFI assets for Nigeria also show a fluctuation but upward trend; increasing

from almost zero percent in 1971 to more than five percent in 2013. South Africa displayed

tremendous growth in NBFI assets increasing from just over one percent to almost 180 percent

over the same period. This is reflective of the potential for NBFIs to becoming sources of

finance for investments in Africa, in the wake of declining credit from traditional banks.

The Figure 4 below shows that after the financial crisis, credit extended to the private sector

by deposit banks for the three countries took a downward turn for all the three countries. Credit

to the private sector by deposit banks for Egypt (PC_E), Nigeria (PC_N) and South Africa

0

20

40

60

80

100

120

140

160

180

200

0

1

2

3

4

5

6

7

8

9

NB

FI a

s a

% o

f G

DP

: So

uth

Afr

ica

NB

FI a

s a

% o

f G

DP

: Eg

ypt

and

Nig

eri

a

NBFI_E NBFI_N NBFI_SA

Global financial crisis

11

(PC_SA) was expressed as a percentage of GDP for each respective country over the period

1971-2013.

Figure 4: Credit to the private sector by deposit banks expressed as percentage of GDP

for Egypt, Nigeria and South Africa: 1971-2013

Source: World Development Indicators (2015), Central Banks of Nigeria, Egypt and South Africa

Figure 4 shows that while deposit bank credit has generally been on an upward trend, after the

year 2008 – the period after the global financial crisis, credit to the private sector has been on

a downward trend. One of the primary reasons is that banks are yet to fully recover from the

financial crisis and also have to adjust to tighter regulatory requirements - mostly emanating

from the Basel III requirements. On the other hand, the trend in NBFI assets has been on an

upward trend over the same period. This clearly demonstrates the emergence of NBFIs as an

important source of long term financing for African countries. Thus, it is necessary to explore

the importance of NBFIs to the finance growth-nexus using evidence from the selected

countries.

3.3. Correlation between NBFIs and economic growth

Before we engage with the empirical assessment of the relationship between NBFIs and

economic growth using econometric technique, we use Figure 5 below to demonstrate the

relationship between the development of NBFIs and economic growth in each country. A

cursory look at Figure 5 suggests some relationship between the growth in NBFIs’ assets and

GDP growth (both in nominal local currency terms), with the relationship appear strongest in

South Africa than the other countries, especially in more recent years.

0

20

40

60

80

100

De

po

sit

ban

k cr

ed

it t

o t

he

p

riva

te s

ect

or

as a

% o

f G

DP

PC_E PC_N PC_SA

Global financial crisis

12

Figure 5: Co-movement of growth in assets of NBFIs and GDP for Egypt, Nigeria and

South Africa: 1971-2013

Source: World Development Indicators (2015), Central Banks of Nigeria, Egypt and South Africa

However, the exact nature of the relationship can only be established with more rigorous

econometric techniques. The following section presents the econometric methods used in the

empirical analysis.

4. METHODOLOGY

The traditional approach to investigating the finance-growth nexus is to regress economic

growth (Y) on financial development (FD) together with other control variables, i.e. Y = f(FD,

control variable(s)) (Levine, 1997). Economic growth is proxied by log of per capita real GDP.

Our decision to use log of per capita real GDP is informed by literature which shows that most

time series studies use the level of per capita GDP while panel/cross countries studies use the

growth rate. As indicated above, FD will be proxied by NBFIs, because it is the objective of

this study to establish the importance of NBFIs in the finance-growth debate.

To empirically test this relationship, this study uses the Johansen cointegration and vector error

correction model within a country specific setting. This approach provides a framework for

testing the existence of a long-run equilibrium relationship between Y and FD and the causality

thereof.

-40%

-30%

-20%

-10%

0%

10%

20%

30%

40%

50%

60%

0%

5%

10%

15%

20%

25%

30%

35%1

971

19

75

19

79

19

83

19

87

19

91

19

95

19

99

20

03

20

07

20

11

% N

BFI

s gr

ow

th

% G

DP

gro

wth

Egypt

Y NBFI Assets

-10%

0%

10%

20%

30%

40%

50%

60%

70%

0%

5%

10%

15%

20%

25%

30%

35%

19

71

19

75

19

79

19

83

19

87

19

91

19

95

19

99

20

03

20

07

20

11

% N

BFI

s gr

ow

th

% G

DP

gro

wth

South Africa

Y NBFI Assets

-100%-50%0%50%100%150%200%250%300%350%400%450%

-20%

0%

20%

40%

60%

80%

100%

120%

140%

19

71

19

73

19

75

19

77

19

79

19

81

19

83

19

85

19

87

19

89

19

91

19

93

19

95

19

97

19

99

20

01

20

03

20

05

20

07

20

09

20

11

20

13

% N

BFI

s gr

ow

th

% G

DP

gro

wth

Nigeria

Y NBFI Assets

13

The multivariate vector error correction model with k lags is expressed as below:

∆Xt = ∏Xt − 1 + ∑ + ┌i ∆Xt − 1 εk t 𝑘

𝑡=1 1

Where Xt = f (Y, FD, control variable) is a 3 x 1 vector. The variables are integrated of order

1, that is I(1). ∆Xt is I(0); ┌i represents 3 x3 short –run coefficient matrices; and εk t shows

normally and independently distributed error terms.

In a trivariate model (where n = 3), the full rank of ∏ matrix and its reduced rank is r. If r = 3,

then the variables Xt are I(0). However, if the rank of the ∏ matrix is zero, then there are no

cointegrating relationships between the variables. The ∏ matrix can also have a reduced rank

in the order of r ≤ (n-1). Thus, in a trivariate model, two reduced ranks are possible: r = 1 (one

cointegrating vector) and r = 2 (two cointegrating vectors).

This paper uses a trivariate model. In the trivariate model one control variable is added at a

time. Y and FD remain constant in the model since they are the variables of interest. Adding

one variable at a time helps to address possible misspecification problem inherent in a bi-

variate model. Furthermore, it helps to test the robustness of the long-run relationship between

Y and FD. Specifically, a trivariate model enables us to test if the long-run relationship is

affected by the control variable used.

In carrying out the analysis, the following steps were followed:

(i) Test for unit root. The study uses augmented Dickey Fuller and the Breakpoint unit

root test.

(ii) If the series are I(1), then we test for cointegration. If no cointegration is established,

then the model is discarded;

(iii) When cointegration exists, weak exogeneity test is carried out to ascertain if at least

one of the variables of interest (Y and FD) is endogenous. If none of the variables

is endogenous, the model is not reported;

(iv) After establishing endogeneity, the model is tested for serial correlation and

heteroscedasticity. If there is serial correlation or heteroscedasticity is found, then

the model is not reported. ;

(v) If the model satisfies the serial correlation and heteroscedasticity, its explanatory

power is also assessed. Specifically only models with an adjusted R2 values greater

than 30% are reported. This is to ensure that the model has a relatively high

explanatory power and good fit.

In the following table we briefly discuss a priori expectation of the control variables based on

theory and literature. All variables were measured as a percentage of nominal GDP except

inflation and interest rates.

14

Table 2: Description of control variables

Concept being

measured

Variable and description

Macroeconomic

stability

Inflation was used as a measure of macroeconomic stability. High inflation rates suggest an

unstable macroeconomic environment, which may reduce investment in productive assets, thus

retarding economic growth.

Interest rates Literature suggests that interest rates influence both the supply and demand of financial

services. High deposit rates are expected to attract savings which are then invested into the

economy. On the other hand, high lending rates and a wide interest rate spread discourage

borrowing for productive assets, thus reducing economic growth.

Government fiscal

policy

Growth in government expenditure could reduce economic growth if it crowds out private

investment. More so, it can also be inflationary, thus reducing economic growth.

Trade openness Exports, imports and openness (exports plus imports as a percentage of GDP). The need to

develop appropriate instruments to finance international trade may facilitate financial

development (Rajan and Zingales, 2000).

Investment Gross fixed capital formation was used as a measure of investment. An increase in investment

is expected to increase the productive capacity of the economy, thus promoting economic

growth.

Natural resources Recent literature suggests that natural resources may inhibit economic growth. This is referred

to as the ‘natural resource curse’. On the other hand, availability of natural resources suggests

potential for industrial growth and exports, both of which are expected to have a positive effect

on economic growth (Sala-i-Martin and Subramanian, 2003).

Infrastructure The availability of infrastructure is expected to facilitate economic transaction. Therefore, the

lack of it may affect economic growth. Also the lack of infrastructure may be indicative of

potential for investment which may stimulate the development of financial instruments to

finance its development. This suggests that the effect of infrastructure on economic growth

and financial development can be positive or negative

Source: Compiled by author

4.1. Data and sources

Data used in estimating the model is annual and covers the period 1971-2013, but was not

available for the entire period for some of the variable. Data was obtained from the World

Development Indicators (WDI), Global Development Finance Database (GDF), International

Financial Statistics (IFS), United Nations Statistics (UN Stats) and Central Banks for Egypt,

Nigeria and South Africa (See Annexure 3 for the full description of variable and period of

coverage). The period of study was chosen solely due to data limitations, which problem is

characteristic of African countries.

15

5. ESTIMATION RESULTS

This section presents and discusses the results. The results are presented in four sections: (i)

results from the unit root test; (ii) cointegration results; (iii) weak exogeneity test results; and

(iv) the long-run relationship between NBFIs and economic growth.

5.1. Unit root test results

In order to test the stationarity of the variables we use two test methods- the Augment Dickey

Fuller and the breakpoint unit root test. The results show that most of the variables are first

difference stationary, that is I(1) series. Hence, it is necessary to carry out cointegration test to

determine whether or not a long-run relationship exist among the variables.

5.2. Cointegration

Cointegration suggests the existence of a long-run equilibrium relationship between economic

variables (Kennedy, 2003; Geda, Ndungu and Zerfu, 2012 and Brooks, 2014). We use the

Johansen cointegration technique to test for cointegration. Table 4 presents the cointegration

test results. In the table ‘K’ indicates the VAR order that produces a white noise residual; and

‘A’ indicates the deterministic trend assumption for each particular model. These are models

that produced meaningful results, that it, they satisfied serial correlation, heteroscedasticity and

adjusted R2 specifications stated above.

Table 3: Cointegration results: Egypt, Nigeria and South Africa

Country Control Variable Observations K A7 Trace Statistic Max Eigenvalue

Egypt

GasRents 36 5 3 36.30[0.00] 26.51[0.01]

Industr 36 5 2 49.14[0.00] 31.23[0.00]

LendingR 32 4 4 55.88[0.00] 32.82[0.01]

NetTaxes 38 3 4 56.74[0.00] 35.23[0.00]

OilRents 38 3 3 41.02[0.00] 26.33[0.01]

OPP 36 5 2 42.72[0.00] 22.62[0.05]

Tel 31 2 2 36.36[0.03] 25.24[0.02]

Tel100 31 2 2 36.98[0.03] 25.90[0.02]

Nigeria GasRents 40 2 4 44.30[0.03] 29.34[0.02]

UrbanPop 38 4 4 46.89[0.02] 28.16[0.02]

South Africa

CPI 38 4 4 59.36[0.00] 38.38[0.00]

DepositR 34 2 4 48.87[0.01] 29.45[0.02]

Exports 39 3 4 48.27[0.01] 31.60[0.01]

GCF 34 8 3 80.40[0.00] 66.64[0.00]

GvtCons 40 2 2 37.58[0.03] 24.47[0.02]

Imports 38 4 3 42.59[0.00] 27.85[0.01]

NetTaxes 37 5 4 50.81[0.01] 26.49[0.04]

Spread 32 4 2 37.18[0.03] 24.25[0.03]

7 Assumption 2 indicates that the level series have no deterministic trend and the cointegrating equations have

intercepts. Assumption 3 indicates that the level series have a linear trend, while cointegrating equations have

intercepts but no trend. Assumption 4 indicates that both level series and cointegrating equations have linear

trends.

16

Notes: See Annexure 2 for definition of control variables. Parentheses [ ] are used to denote the relevant

probability values.

Source: Compiled by author

Of the 22 models estimated for each country, the trace test and the max eigenvalue test show

that there is cointegration between NBFIs and economic growth in 8 models each for Egypt

and South Africa. Only 2 models show evidence of cointegration between these variables in

Nigeria. This clearly shows that, in Nigeria there is weak relationship between NBFIs

development and economic growth. This is not surprising given the low level of development

of the NBFIs in Nigeria. In contrast, in Egypt and South Africa there appears to be a strong

relationship between the NBFIs and economic growth. Next we consider the weak exogeneity

test results to determine the nature (direction) of causal link between the two variables. The

existence of cointegration implies that causality must at least run from one of the variables to

the other.

5.2.1. Causality between NBFIs and economic growth

A cross-reading of literature suggests that the relationship between financial development

(using various indicators) and economic growth is by no means simple and clear cut. The first

strand of literature suggests that financial development leads to economic growth by facilitating

mobilisation of savings and efficient allocation of capital (Levine, 2004). Therefore, the apriori

expectation is that the development of NBFIs (in this case) positively influences economic

growth. However, this proposition was also criticised on the basis that financial development

eliminates liquidity constraints, and thus reduces the incentive to save, which may further

inhibit economic growth.

The second hypothesis is that there is a sequential two-way causality between financial

development and economic growth (Patrick, 1966). Patrick argued that the causality can either

be demand-following or supply-leading. A demand-following scenario arises when causality

runs from economic growth to financial development. In this case, NBFIs should develop in

response to the demand for its services by the real economy. A supply-leading scenario arises

when causality runs from financial development to economic growth. Thus, the supply of

services by NBFIs is in advance to the demand for such by the real economy.

Lastly, the causality between financial development and economic growth can also be a

simultaneous two-way causality. The simultaneous two-way causality can either be a vicious

or virtuous cycle (Berthelemy and Varoudakis, 1996). A vicious cycle occurs when economic

growth is too low such that it prevents the development of the financial sector, which in turn

prevents economic growth. In this instance, the long-run coefficient of NBFIs on economic

growth (or vice-versa) is expected to be negative. On the other hand, a virtuous cycle arises

when a high level of economic growth supports the development of the financial sector, which

in turn stimulates further economic growth. Thus, the long-run coefficient of NBFIs on

economic growth (or vice-versa) is expected to be positive.

17

Given the conflicting views in respect of the causal link between financial development and

economic growth, what prevails within each particular setting becomes an empirical issue. In

this paper we use the weak exogeneity approach to test the direction of causality between

NBFIs and economic growth. The nature of causality (whether positive or negative) is

presented in Table 5 in the section below. The weak exogeneity results and the direction of

causality thereof are reported in Table 4. The table reports the Chi-square statistic and the

probability value of the test. Specifically, it presents three null hypotheses: (i) the two-way

causality between economic growth and NBFIs, (ii) causality running from economic growth

to NBFIs, and (iii) causality running from NBFIs to economic growth. A ‘Yes’ indicates that

the null hypothesis cannot be rejected while a ‘No’ indicates that the null hypothesis is rejected.

The results are then discussed separately for each country.

Table 4: Weak exogeneity Test for Egypt, Nigeria and South Africa

Country

Control Variable

Weak Exogeneity Test

Causality between Y and FD

Null Hypothesis

Y FD Y↔FD Y→FD Y←FD

Egypt GasRents 17.10[0.00] 5.53[0.02] Yes Yes Yes

Industr 16.27[0.00] 3.06[0.08] Yes Yes Yes

LendingR 4.12[0.04] 9.69[0.02] Yes Yes Yes

NetTaxes 3.56[0.06] 3.31[0.07] Yes Yes Yes

OilRents 3.81 [0.05] 1.22[0.27] No No Yes

OPP 4.69[0.03] 0.02 [0.88] No No Yes

Tel 16.26 [0.00] 12.63[0.00] Yes Yes Yes

Tel100 16.25[0.00] 12.42 [0.00] Yes Yes Yes

Nigeria GasRents 2.70 [0.10] 5.26 [0.02] No Yes No

UrbanPop 6.56[0.01] 1.55[0.21] No No Yes

South Africa CPI 5.12 [0.02] 0.31[0.58] No No Yes

DepositR 5.62 [0.02] 0.27[0.60] No No Yes

Exports 4.10 [0.04] 0.51[0.48] No No Yes

GCF 3.16 [0.08] 14.50[0.00] Yes Yes Yes

GvtCons 5.11 [0.02] 0.01 [0.92] No No Yes

Imports 3.55[0.06] 0.55[0.46] No No Yes

NetTaxes 3.00[0.08] 4.97 [0.03] Yes Yes Yes

Spread 2.95[0.09] 0.54[0.46] No No Yes

Notes: See Annexure 2 for definition of control variables. Parentheses [ ] are used to denote the relevant

probability values. ‘Yes’ indicates the null hypothesis cannot be rejected while a ‘No’ indicates that null

hypothesis in rejected. Y – economic growth and FD - NBFIs

Source: Compiled by author

Egypt

Results presented in the Table 4 show a predominantly two-way causality between economic

growth and NBFIs in Egypt. Only two models reported causality running from NBFIs to

economic growth only without a reverse causality. This suggests that, in Egypt the relationship

18

between NBFIs and economic growth would be virtuous cycle if the two systems positively

and significantly influence each other. In this case, the long-run coefficients are expected to be

positive. However, if the long-run coefficients are negative, this implies that the relationship

between NBFIs and economic growth in Egypt follows a vicious cycle. Thus, the low level of

economic growth leads to under-development of NBFIs, which in turn hinders economic

growth.

Nigeria

With regards to Nigeria, evidence of causality between NBFIs and economic growth is not only

weak, but mixed. One of the two models reported show that the direction of causality runs from

economic growth to NBFIs, while the other one shows causality in the opposite direction. As

indicated above, the weak results for Nigeria could be due to the fact that NBFIs are still under-

developed. The under-development of the NBFIs suggests that the financial system is not

capable of providing products and instruments required by the real economy to efficiently

facilitate allocation of capital to productive units.

South Africa

The weak exogeneity test results presented in Table 4 show that causality between NBFIs and

economic growth in South Africa pre-dominantly runs from NBFIs to economic growth. This

suggests that the relationship is likely to be supply-leading, whereby the NBFIs creates

financial products required real economy in advance, thus facilitating economic growth. These

results are not surprising given the level of development of NBFIs in South Africa; which is

the most developed in Africa and comparable to developed countries. Well-developed NBFIs

are expected to mobilise savings, provide mechanisms for risk management and the efficient

allocation of capital, thus enhancing economic growth.

The next question that follows is: What is the nature (direction and significance) of the causality

between NBFIs and economic growth? We present our findings in respect of the nature of

causality in the following section.

5.2.2. Long-run relationship between NBFIs and economic growth

Once the direction of causality has been established using the weak exogeneity test, the next

task is to assess the nature (whether positive or negative and the economic significance) of the

long-run relationship that exists between NBFIs and economic growth. If causality runs from

NBFIs to economic growth, it means economic growth is endogenous. We then normalise on

economic growth in order to obtain the long-run coefficients (elasticities) of NBFIs on

economic growth. If the long-run coefficients are positive and significant, this is in the right

direction. Thus, policies must be implemented to strengthen the development and efficiency of

the financial sector.

On the other hand, if NBFIs is endogenous, we condition the model to obtain the long-run

coefficients of economic growth on the development of NBFIs. Positive and economically

19

significant coefficients suggest that the growth of the economy lead to the development of the

NBFIs.

However, if the evidence showing the nature of the relationship is weak, it suggests that factors

beyond NBFIs and economic growth are at play. In this case, there is urgent need to create an

enabling environment to support the development of NBFIs which in turn, stimulate economic

growth.

The long-run coefficients are presented in Table 5 and the table also reports the coefficients of

error correction term. The coefficients of the ECM describes the speed of adjustment by the

dependent variable back to equilibrium, after a shock in the short run. When economic growth

is endogenous in the model, a low ECM suggests inefficiencies of the NBFIs in facilitating

economic growth. Thus, the focus on policy intervention must be to improve efficiency of the

NBFIs.

On the other hand, if the ECM is low when NBFI in endogenous in the model, it suggest

rigidities within the economy. Therefore, it must the objective of relevant policy interventions

to reduce rigidities existent within the economy.

The long-run coefficients and the ECM for NBFIs and economic growth are reported separately

for the two scenarios: (i) where causality runs from NBFIs to economic growth; and (ii) where

causality runs from economic growth to NBFIs.

The long-run coefficients and ECM are presented in Table 5 below. The third column in the

table presents the coefficient and its corresponding t-statistic, while the fourth column reports

the ECM coefficients. The last column reports the economic significance of the coefficients.

Long-run relationship between NBFIs and economic growth when economic growth is

endogenous

This section presents the long-run parameters and the ECM when causality runs from NBFIs

to economic growth. In this case, economic growth is normalised on (conditioned for) in the

model in order to get the elasticity of economic growth to changes in NBFIs. We present the

results in Table 5 below.

Table 5: Long-run parameters of models with causality running from NBFIs to economic

growth

Country Control Variable LR coefficient of FD on Y ECM Y←FD (Sing of Effect)

Egypt OilRents 3.41 [6.70] -0.01[ 1.84] Positive Significant

OPP 2.52 [5.47] -0.03[ 3.69] Positive Significant

GasRents -0.05[-0.19] -0.05[2.87] Negative Not Significant

Industr 0.36[5.57] -0.12[-4.13] Positive Significant

LendingR -0.91[-1.38] -0.01[-1.84] Negative Not Significant

NetTaxes 0.01[0.31] -0.25[-3.80] Positive Not Significant

Tel 0.12[2.39] -0.18[-4.47] Positive Significant

Tel100 0.12[2.20] -0.16[-4.45] Positive Significant

20

Nigeria UrbanPop 0.21[2.34] -0.37 [-3.70] Positive Significant

South

Africa

CPI -0.00 [-0.12] -0.51 [-2.48] Negative Not Significant

DepositR 0.18[1.24] -0.15 [-4.45] Positive Not Significant

Exports 0.48[3.78] -0.16 [-3.52] Positive Significant

GvtCons 0.87 [3.10] -0.02[-2.97] Positive Significant

Imports 0.04[1.62] -0.21[-2.53] Positive Significant

Spread -0.49[-2.25] -0.02[ 2.57] Negative Significant

GCF 0.03[0.30] -0.10[1.18] Positive Not Significant

NetTaxes -65.01[-3.76] -0.00[1.79] Negative Significant

Notes: See Annexure 2 for definition of control variables. Parentheses [ ] are used to denote the relevant

probability values. Y – Economic growth and FD - NBFIs

Source: Compiled by author

Egypt

Of the eight models reported for Egypt, six of them show a positive relationship between NBFIs

and economic growth. Four of the six models with a positive effect show that the influence of

NBFIs on economic growth is significant, while the remaining two show that the effect is not

significant. The other two models that reported a negative effect of NBFIs on economic growth

show that such effect is not significant. The long-run coefficients of NBFIs on economic

growth range from 0.01 to more than 3.4. This shows that the elasticity of economic growth in

response to changes in NBFIs is significant depending on the conditioned factors. Overall, the

weight of the evidence shows that the effect of NBFIs on economic growth in Egypt is positive

and significant.

In respect of the models which reported a positive effect on NBFIs on economic growth, four

models have the coefficients of the error correction term ranging from 0.12 to 0.25. The

remaining two models with a positive effect have the coefficients of the error correction term

of 0.01 and 0.03. On the other hand, those models that reported a negative effect on NBFIs on

economic growth have the coefficients of the error correction term between of 0.01 and 0.05.

If we go by the significant and positive coefficients of the error correction term, it shows that

on average about 10% of disequilibria is corrected in a year, thus it would take closed to 10

year for full equilibrium to be restored after a shock. This evidence suggests that the efficiency

of NBFIs in Egypt is relatively low.

Nigeria

Although the evidence of a relationship between NBFIs and economic growth is weak for

Nigeria, the only model reported in Table 5 above shows positive relationship. The model

shows that the elasticity of economic growth to changes in NBFIs is 0.21 and the ECM is 0.37.

These results are encouraging.

South Africa

21

Of the eight models reported for South Africa, five show a positive effect of NBFIs on

economic growth, while three show a negative effect. Three of the five models that reported a

positive relationship, show that the effect of NBFIs on economic growth is significant, while

the other two show that the effect is not significant. On the other hand, of the three models that

reported a negative effect, two show that the effect of NBFIs on economic growth is significant,

and one model reporting an unusually high elasticity of more than one (65) with a zero

coefficient of error correction term. If we discount the model with the extreme elasticity, the

weight of the evidence would suggest that the effect of NBFIs on economic growth in South

Africa is moderately positive.

The ECM for South Africa range between 0.02 and 0.16 for models that reported a positive

relationship between NBFIs and economic growth. This shows that, there is need to improve

the efficiency of the NBFIs, if South Africa is to fully reap the economic benefits emanating

from the development of NBFIs.

Long-run relationship between NBFIs and economic growth when NBFI is endogenous

This section presents the long-run parameters and the ECM when causality runs from economic

growth to NBFIs. In this case, NBFIs is normalised on (conditioned for) in the model in order

to get its elasticity to changes in economic growth. We present the results in Table 6 below.

Table 6: Long-run parameters of models with causality running from economic growth

to NBFIs

Country Control Variable

LR coefficient of

Y on FD ECM Y→FD (Sign of Effect)

Egypt

GasRents -20.65[-3.14] -0.10[1.97] Negative Significant

Industr 2.78[12.60] -0.11[-1.61] Positive Significant

LendingR -1.10[-0.23] -0.09[-3.19] Negative Not Significant

NetTaxes 69.91[4.60] -0.01[-2.44] Positive Significant

Tel 8.12[5.77] -0.11[-3.77] Positive Significant

Tel100 8.07[5.72] -0.10[-3.75] Positive Significant

Nigeria GasRents 0.74[0.50] -0.22[-3.80] Positive Not Significant

South

Africa

GCF 42.60[2.27] 0.02[3.05] Positive Significant

NetTaxes -0.02[-0.05] 0.61[2.68] Negative Not Significant

Notes: See Annexure 2 for definition of control variables. Parentheses [ ] are used to denote the relevant

probability values. Y – Economic growth and FD - NBFIs

Source: Compiled by author

Egypt

Four of the six models reported for Egypt show that economic growth positively influence the

development of NBFIs, and significantly so. On the other hand, the remaining two models

show a negative effect of economic growth on NBFIs, of which one model shows that the effect

is significant. The elasticities of NBFIs to changes in economic growth are more than one for

22

all the six models reported. This shows that growing the economy will generate more growth

in the financial sector, which should in turn influence economic growth as demonstrated above.

In addition, the results show that the ECM for Egypt is between 0.01 and 0.11. This is reflective

of possible rigidities within the Egyptian economy. Thus, in order to realise greater

development of the financial sector as the economy grows, policies and frameworks should be

in place to minimise rigidities that may exist within the economy.

Nigeria

Although results presented in Table 6 above show that the effect on economic growth on the

development of NBFIs is positive, such effect is not significant and the evidence is weak. The

elasticity of NBFIs to changes in economic growth is 0.74 and the ECM is 0.22.

South Africa

When NBFIs is endogenous in the model, the effect of economic growth on the development

of NBFIs in South Africa is mixed. In addition, such evidence is weak, as only two models

were reported in this case (when causality runs from economic growth to NBFIs). This shows

that the development of NBFIs in South Africa is pre-dominantly supply-leading.

6. SUMMARY OF RESULTS AND CONCLUSIONS

The following key findings emerge from this paper: Firstly, cointegration tests reported in this

paper show the existence of a long-run relationship between NBFIs and economic growth in

Egypt and South Africa. The evidence for such a relationship in Nigeria is weak, as only two

models (out of a possible 22) were reported. These studies are in line with previous empirical

studies which also found that countries with relatively more developed financial systems

exhibit evidence of a long-run relationship between NBFIs and economic growth (Haiss and

Sumegi, 2008; Cheng and Degryse, 2010; and Meng and Pfau, 2010).

Secondly, the direction of causality between NBFIs and economic growth in Egypt and South

Africa predominantly runs from NBFIs to economic growth and is positive and significant.

Thus, financial development in these countries is supply-leading. These results are not

surprising given the level of development of NBFIs in South Africa; which is the most

developed in Africa and comparable to developed countries. Well-developed NBFIs are

expected to mobilise savings, provide mechanisms for risk management and the efficient

allocation of capital, thus enhancing economic growth.

Thirdly, the coefficients of the error correction term for all the three countries suggests

inefficiencies in the financial system and rigidities within the economies. Specifically, the

ECM reported very low values (sometimes close to zero) which suggest that shocks emanating

from either the economy or the financial system may not be quickly corrected to restore full

equilibrium.

23

Fourthly, except for Egypt, when causality runs from economic growth to NBFIs, the

relationship between NBFIs and economic growth is weak and mixed. Evidence on Egypt also

shows that growth of the economy positively influences the development of NBFIs. This

suggests that the relationship is also demand-following, though not as strong as supply-leading.

On the basis of the empirical results discussed above, we recommend the following for each of

the countries and other African countries:

i. Egypt: Policies be developed and implemented to improve efficiency of the

financial system. This will enable the financial system to efficiently carry out its

intermediation role. More efficient intermediation is likely to enhance economic

growth, given the existence of significant positive causal link between NBFIs and

economic growth.

ii. Nigeria: Weak evidence of causality between NBFIs and economic growth in

Nigeria suggest that factors exogenous to the finance-growth nexus may be at play.

In this regard, sufficient policies must be developed to create an enabling

environment which supports the development of the financial sector and growth of

the economy. Once growth of the economy reaches a certain threshold, both

systems may start reinforcing one another resulting in a positive relationship.

iii. South Africa: The World Bank Financial Inclusion Data (Findex, 2014) show that

South Africans are the most borrowed and this may pose an economic and financial

threat if bankruptcy becomes significant. For this reason, the relevant authorities

must deliberately implement policies to stimulate economic growth and also

improve efficiency of the financial system. By doing so, this may create real

demand for financial instruments, thus further stimulating financial development.

iv. Lastly, to other African and emerging market economies, we recommend that

authorities deliberately implement policies and reform programmes to stimulate

development of NBFIs and also improve their efficiency in intermediating funds.

This will unlock funds for investment in long-term productive assets.

The results presented above contribute to the finance-growth literature in two ways (i) by

providing country specific time-series evidence to support the finance-growth nexus using

NBFIs; and (ii) by providing evidence on causality between NBFIs and economic growth in

Africa. Given that traditional bank lending has declined while NBFIs have emerged as a

potential source of long-term capital, results from this study should be invaluable to policy

formulation pertaining to the importance and development of NBFIs in Africa and other less

developed countries.

24

REFERENCES Ahmed, A. and Wahid A., (2010). “Financial structure and economic growth link in African countries:

a panel cointegration analysis”. Journal of Economic Studies Vol. 38 No. 3, 2011

Aldermen, H. and Yemtsov, R., (2013). “How Can Safety Nets Contribute to Economic Growth?”.

World Bank Policy Research Working Paper 6437

Ang, J., (2008). “What are the mechanisms linking financial development and economic growth in

Malaysia?” Economic Modelling 25 (2008) 38–53

Arestisa, P., Luintelc, A. and Luintel, K., (2010). “Financial structure and economic growth: evidence

from time series analyses”. Applied Financial Economics, 20, 1479–1492

Aziakpono, M., (2011). “Financial development and economic growth: theory and a survey of

evidence”. J.Stud.Econ.Econometrics, 35(1)

Bagehot, W., (1874) “Lombard Street: A description of Money Market”. Scribner, Armstrong and Co.,

New York

Brooks, C., (2014). “Introductory Econometrics for Finance”, 3rd Edition. Cambridge University Press

Casey, B. and Dostal, J., (2008). “Pension Reforms in Nigeria: How not to Learn from Others”. Global

Social Policy 1468-0181 vol. 8(2): 238–266; 090640

Čihák, M., Demirgüč-Kunt, A., Feyen, E. and Levine, R., (2013). “Financial Development in 205

Economies, 1960 to 2010”. NBER Working Paper Series 18946

Cheng and Degryse (2010) “The impact of bank and non-bank financial institutions on Local Economic

growth in China”. BOFIT Discussion Papers 22/2007

Davies, E. and Hu Y., (2008). “Does funding of pensions stimulate economic growth?” Journal of

Pension Economics and Finance / Volume 7 / Issue 02, pp 221 - 249

Eschenbach, F., (2004). “Finance and Growth: A survey of evidence”. Tibergen Institute TI 2004-039/2

Financial Stability Board (2014). “Global Shadow Banking Report 2014”

Fischer, I., (1933). “The Debt - Deflation Theory of Great Depressions”. Econometrica, pp 337-357

Geda, A., Ndungu, N. and Zerfu, D., (2012). Applied Time Series Econometrics: A practical guide to

macroeconomic researchers with a focus on Africa. University of Nairobi Press

Gertler, M., (1988). "Financial Structure and Aggregate Economic Activity: An Overview." Journal of

Money, Credit and Banking vol. 20, no. 3, pp. 559-88

Goldsmith, R., (1959). “Financial Structure and Development, a subject for International Comparative

Study”. NBER, pp 114 - 123

Gurley, J. and Shaw, E., (1955). “Financial Aspects of Economic Development”. The American

Economic Review, Vol. 45, No. 4, pp. 515-538

Haiss, P. and Sumegi, K., (2008). “The relationship between insurance and economic growth in Europe:

a theoretical and empirical analysis”. Empirica 35:405–431

Harris, L., 2012. From Financial Development to Economic Growth and Vice Versa: A Review of

International Experience and Policy Lessons for Africa. Journal of African Economies, Vol. 21,

AERC Supplement 1, pp. i89–i106

25

Holzmann, R., (1997). “Pension reform, financial market development and economic growth:

Preliminary evidence from Chile”. Staff Papers (International Monetary Fund), Vol. 44, No. 2,

pp. 149-178

Kennedy, P., (2003).”A Guide to Econometrics”, 5th Edition. Blackwell Publishing

King R. and Levine R., (1993a). “Finance and Growth: Schumpeter might be right”. The Quarterly

Journal of Economics, pp 717-736

King R. and Levine R., (1993b). “Finance Entrepreneurship and Growth: Theory and Evidence”.

Journal of Monetary Economics 32 513-542.

KPMG Africa Limited (2013). “Financial Services in Africa”.

Levine, R., (1997). “Financial Development and Economic Growth: views and agenda” Journal of

Economic Literature, pp 688-726

Levine, R., (2004). “Finance and Growth: theory and evidence”. NBER Working Paper Series 10777

Liang, H. and Reichert A., (2015). “The impact of banks and non-bank financial institutions on

economic growth” The Service Industries Journal Vol. 32, No. 5, 699–717

Mckinnon, R., (1973). “Money and Capital in Economic Development”. The Brookings Institution,

McGill-Queen's University Press

Meng, C. and Pfau, W., (2010). “Role of pension funds in capital market development”. GRIPS

Discussion Paper 10-17

Mishkin, F., (2007). “The economics of money, banking and financial markets”, 8th Edition, Pearson

Education

Mo Ibrahim Foundation (2013). “Africa Ahead: The Next Fifty Years”.

Nassr, I. and Wahinger G., (2014). “Non-bank debt financing for SMEs: The role of securitisation,

private placements and bonds”. OECD Journal: Financial Market Trends Volume 2014/1

Ndugbu, M., at al (2015). “Bank And Non-Bank Financial Institutions And The Development Of The

Nigerian Economy”. International Journal for Innovation Education and Research Vol:-3 No-8

Osuala, A. and Odunze, C., (2014). “Do Non-Bank Financial Institutions’ Loans and Advances

Influence Economic Growth? A Bounds Test Investigation”. European Journal of Business and

Management Vol.6, No.27

Pagano, M., (1993). “Financial Markets and Growth: An overview”. European Economic Review, vol.

37, issue 2-3, pages 613-622

Patrick H., (1966). “Financial Development and Economic Growth in underdeveloped countries”.

Economic Development and Cultural Change, Vol. 14, No. 2, pp. 174-189

Rajan, R. and Zingales, L., (2000). “The great reversals: the politics of financial development in the

20th century”. OECD Economics Department Working Papers No. 265

Sala-i-Martin, X. and Subramanian, A., (2003). “Addressing the natural resource curse: An illustration

from Nigeria”. International Monetary Fund

Shaw, E., (1973). “Financial Deepening in Economic Development”. Oxford University Press

Stiglitz, J., (1994). “Economic Growth Revisited,” Industrial and Corporate Change, 3(1), pp. 65-110.

26

Stiglitz, J., (2000). “Liberalization, Moral Hazard in Banking and Prudential Regulation: Are Capital

Requirements Enough?” American Economic Review, 90(1), March 2000, pp. 147-165.

Sufian, F. and Majid, M., (2009). “Post-Crisis productivity Change in Non- Bank Financial Institutions:

Efficiency Increase or Technological Progress?”. Journal of Transnational Management,

14:124–154

United Nations Economic Commission for Africa (2015). “Illicit Financial Flows: Report of the High

Level Panel on Illicit Financial Flows in Africa”

Vittas, D., (1997) The Role of Non-Bank Financial Intermediaries in Egypt and other MENA countries

Development Research Group, World Bank

World Bank (2013). “Rethinking the role of State in Finance”. Global Financial Report 2013.

World Bank (2013). “Financing for Development Post 2015”

27

ANNEXURES Annexure 1: Growth in assets of NBFI across the world: 2002 – 2013

US$ trillion 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013

%

Growth

Insurance

Companies 13,7 16,6 18,8 20 21,9 24,1 21,3 22,9 24,6 25,7 26,6 27,4 99%

Pension Funds 11,7 13,4 14,9 15,9 17,6 18,9 18,6 21 23,4 24,2 25,9 27,4 135%

Other

Financial

Intermediaries 26,4 32,4 38,1 44 51,8 61,5 58,5 60,2 65,3 67,5 70,5 75,2 185%

Source: Global Shadow Banking Monitoring Report 2014

Annexure 2: NBFI Assets as a percentage of GDP for the world average, high income

countries and sub-Saharan Africa countries: 1970-2011

Source: Global Development Finance Database (2013)

0

5

10

15

20

25

NB

FI A

sse

ts a

s a

% o

f G

DP

World Average High Income SSA

28

Annexure 3: Description of variables

Country

Variable Definition Egypt Nigeria

South

Africa

Y Real GDP per capita 1971-2013 1971-2013 1971-2013

NFBI

Total assets held by financial institutions that do not accept transferable deposits but that

perform financial intermediation by accepting other types of deposits or by issuing securities or

other liabilities that are close substitutes for deposits as a % of GDP. 1971-2012 1971-2013 1971-2013

Agric

Agriculture corresponds to ISIC divisions 1-5 and includes forestry, hunting, and fishing, as

well as cultivation of crops and livestock production. 1971-2013 1981-2013 1971-2013

CPI Consumer Price Index (2010 as base year) 1971-2013 1971-2013 1971-2013

DepositR Deposit interest rate is the rate offered by commercial banks on three-month deposits. 1976-2013 1971-2013 1977-2013

Elec

Electric power consumption measures the production of power plants and combined heat and

power plants less transmission, distribution, and transformation losses and own use by heat and

power plants. 1971-2011 1971-2011 1971-2011

ElecPecapita Electric power consumption per capita 1971-2011 1971-2011 1971-2011

ERav National Currency per U.S. Dollar, period average 1971-2013 1971-2013 1971-2013

ERend National Currency per U.S. Dollar, end of period 1971-2013 1971-2013 1971-2013

Exports

Exports of goods and services represent the value of all goods and other market services

provided to the rest of the world 1971-2013 1971-2013 1971-2013

GasRent

Natural gas rents are the difference between the value of natural gas production at world prices

and total costs of production. 1971-2013 1971-2013 1971-2013

GCF

Gross capital formation (formerly gross domestic investment) consists of outlays on additions

to the fixed assets of the economy plus net changes in the level of inventories 1971-2013 1971-2013 1971-2013

GvtCons

Government consumption - final consumption expenditure includes all government current

expenditures for purchases of goods and services (including compensation of employees).

1971-2013 1971-2013 1971-2013

Imports

Imports of goods and services represent the value of all goods and other market services

received from the rest of the world. 1971-2013 1971-2013 1971-2013

Industri

It comprises value added in mining, manufacturing, construction, electricity, water, and gas

1971-2013 1981-2013 1971-2013

LendingR Lending rate is the rate charged by banks on loans to the private sector. 1976-2013 1971-2013 1971-2013

Mrents

Mineral rents are the difference between the value of production for a stock of minerals at

world prices and their total costs of production. 1971-2013 1971-2013 1971-2013

29

NEREExRate Nominal Effective Exchange Rate 1979-2013 1979-2013

NetTaxes

Net taxes on products (net indirect taxes) are the sum of product taxes less subsidies. Product

taxes are those taxes payable by producers that relate to the production, sale, purchase or use of

the goods and services. 1971-2013 1981-2013 1971-2013

OilRent

Oil rents - the difference between the value of crude oil production at world prices and total

costs of production as a percentage of GDP. 1971-2013 1971-2013 1971-2013

OPP

Openness to trade is the total value of exports and imports, expressed as a percentage of GDP

1971-2013 1971-2013 1971-2013

PDensity Population density is midyear population divided by land area in square kilometers. 1971-2013 1971-2013 1971-2013

Pop Number of people in a particular country 1971-2013 1971-2013 1971-2013

Spread It is the difference between the deposit and the lending rates. 1976-2013 1971-2013 1977-2013

Tel

Telephone lines are fixed telephone lines that connect a subscriber's terminal equipment to the

public switched telephone network and that have a port on a telephone exchange. Integrated

services digital network channels and fixed wireless subscribers are included 1975-2013 1981-2013 1971-2013

Tel100 Telephone lines per every 100 people. 1975-2013 1981-2013 1971-2013

UrbanPop

Population in urban agglomerations of more than one million is the country's population living

in metropolitan areas that in 2000 had a population of more than one million people. 1971-2013 1971-2013 1971-2013

30

Annexure 4: NBFIs in Egypt, Nigeria and South Africa

Egypt

NBFIs in Egypt are regulated under the Regulation of Non-banking Financial Markets and

Instruments Law 10 of 2009. The law provided for the establishment of the Egyptian Financial

Supervisory Authority (EFSA), which authority is responsible for the regulation and

supervision of NBFIs in Egypt. Specifically, the EFSA is responsible for the following

activities, inter ilia:

‘License non-banking financial activities;

Inspect entities licensed to operate in non-banking financial activities;

Ensure transparency and competitiveness in rendering non-banking financial services

through the proper control of non-banking financial markets;

Protect non-banking market dealers' rights and take necessary measures to minimize

market manipulations and fraud, taking into heed the potential commercial risks of

dealing in these markets; and

Supervise training of non-banking market staffs and enhance their efficiency’.

NBFIs under the supervision of EFSA include capital markets, life insurance, short-term

insurance, mortgage finance, leasing, factoring and securitisation. We review some of these

below.

Short term insurance in Egypt

The insurance industry in Egypt started late in the 19th century. The first insurance company

(the National Insurance Company) in Egypt was established in 1900. This was followed by Al-

Sharq and Misr for General Insurance in 1933 and 1934 respectively. During the nationalisation

programme under Law No 23 of 1957, all insurance companies were nationalised. Following

the merger decree of 1964, the number of insurance companies was reduced to four. In 1975,

after the 1973 war, new laws were enacted to allow private participation in the insurance sector.

In 1981, market liberalisation policies allowed 49% foreign ownership, which percentage was

increased to 100% in 1998. Currently, there are 28 insurance companies registered with EFSA.

Pension Funds in Egypt

The purpose of pension funds is to provide instruments to minimise risk of loss of income as a

result of termination of employment, retirement, accidents or incapacity due to disease. In

2013, there were 4.7 million subscribers contributing more than US$660 million per year. Total

assets and reserves of pension funds in Egypt amounted to more than US$6.5 billion in 2013.

8This law was enacted as a replacement of the the Egyptian Insurance Supervisory Authority, the Capital Market Authority, and the Mortgage Finance Authority in enforcing the provisions of the Insurance Supervision and Control Law no.10 of 1981, the Capital Market Law no.95 of 1992, the Depository and Central Registry Law no.93 of 2000, the Mortgage Finance Law no.148 of 2001, as well as any other related laws and decrees that are part of the mandates of the above authorities. 9Capital market refers to stock markets which are not subject of investigation in this paper.

31

As a show of the importance of NBFIs to financing investment, pension funds in Egypt

financed more than US$5.8 billion worth of investment.

South Africa

South Africa’s boast of one of the fastest growing and unusually large NBFIs sector compared

to other emerging countries (IMF, 2014). Currently NBFIs account for more than two thirds of

financial sector assets. Specifically, South African Reserve Bank Data (2015) show that assets

of NBFIs grew from US$140 billion in 2001 to more than US$650 billion in 2014.

Regulation of NBFIs in South Africa currently falls under the auspices of the Financial Services

Board (FSB). However, the country is in the process of regulatory reform, to adopt a new model

of regulation call the ‘Twin-Peak Model’. In terms of this new model, regulation of certain

functions of the financial institutions will be split across different regulatory bodies. For

instance, prudential regulation which is required to ensure safety, stability and solvency of

financial institutions will fall under one regulatory body, while market conduct regulation

which focuses on consumer protection will be done by a different regulator National Treasury

SA, 2011).

The NBFI sector in South Africa includes financial institutions such as insurance companies,

pension and provident funds, Public Investment Corporation, collective investment schemes

and development financial institutions. We provide a brief over of these institutions below.

Insurance companies

The insurance sector comprise of short- and long-term insurance and re-insurance. Assets

currently held by the insurance sector in South Africa account for almost a quarter of financial

sector assets in the country. Currently there are 87 short-term insurers, 78 long-term insurers

and 6 reinsurers (IMF, 2014). The insurance sector in South Africa has a number of

intermediaries, with more than 10 000 registered financial service providers.

Pension and provident funds

In South Africa, 60% of pension funds are privately funded and administered (Brown, 2013).

Although the country does not have a compulsory pension scheme, the Government operates a

pension fund for its employees, namely the Government Employees Pension Fund (GEPF).

The mandate of the DEPF is to administer and manage pensions and retirement benefits are

Government employees. The GEPF is estimated to have more than 1.2 million members and

has more than US$100 billion worth of assets.

Public Investment Corporation (PIC)

The PIC is a registered financial services provider, established in 1911 and wholly owned by

Government. It manages funds on behalf of public entities, and its main client is the GEPF

which makes up more than 90 percent of assets under its management. The PIC is one of the

32

largest investment asset managers in Africa, with more than US$130 billion under its

management.