1. The Impact of Monetary Policy on Bank Balance Sheet Variables in
sub-Saharan Africa: Evidence for a Bank Lending Channel
MASTER OF ARTS ECONOMICS THESIS
BY
MPHATSO ELIAS ACKIM
BSoc. Sc. (Economics)
June 17th 2016
1

2. OUTLINE OF PRESENTATION
• Background Information
• Problem Statement and Justification
• Objectives and Hypotheses
• Methodology
• Empirical Results and Interpretation
• Conclusion and Policy Implications
2

3. BACKGROUND INFORMATION
Effectiveness of Monetary Policy
• Underlying transmission mechanisms of
monetary policy remains a subject of
debate.
• The Bank Lending Channel is one of the
transmission mechanisms that
recognizes the role played by
commercial banks.
• Modigliani - Miller Theorem argues that
banks are not affected by problem of less
Banking Systems of SSA
• Depth of financial system is low
• Banks dominate formal financial
systems.
• Bank credit dominates external
finance of private sector firms.
• Low Credit supply - Assets are held
in the form of government
securities & liquid assets.
3

4. PROBLEM STATEMENT AND JUSTIFICATION
4
0.00
20.00
40.00
60.00
80.00
100.00
120.00
140.00
160.00
180.00
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Credit(asaPercentageofGDP)
Time Period
Trends in Bank Credit to Private Sector (2000 -2014)
SSA OECD East Asia and Pacific Latin America Middle East

5. PROBLEM STATEMENT AND JUSTIFICATION
• Have mostly examined effects of monetary policy on credit
supply, but not on deposits and liquid assets.
• {Matemilola et al. (2014); Amidu (2014); Nana and
Samson (2014) ; & Djiogap & Ngomsi (2012)}
Known Studies
• Are of closed economy type.
• The effects of international reserve accumulation has not
been examined as suggested by Shrestha (2013) in a study
of East Asia
The models used
• The effects of monetary policy working through its
interaction with levels of capitalization have less been
examined.
• Only Walker (2012); Kabiro & Nyamongo (2014), have
done this but their analyses are confined to EAC countries
Modigliani – Miller
Theorem
5

6. PROBLEM STATEMENT AND JUSTIFICATION CONTI… 6
• Needs to investigate all three main variables in the
balance sheets of commercial banks
• The variables are interrelated and changes in any of
them affects the whole portfolio (Shrestha 2013).
Meaningful Analysis
•Analysis of balance sheets is important for financial
stability
•Commercial banks generates boom and bust cycles in
the economy, through expansion and contraction of
credit flows {Villar (2006); Adrian and Shin (2009);
Brunnermeier et al. (2009); Mittnik and Semmler
(2011)}
•Weaknesses in balance sheets variables can ignite and
propagate financial crises (Allen et al., 2002).
Justification

7. OBJECTIVES OF THE STUDY
• Examine the effects of Monetary Policy on balance
sheet variables of commercial banks
The Main Objective
•Examine the effects of real interest rates on balance
sheet variables of commercial banks.
•Investigates the interaction effect of real interest rates
and capitalization ratios on the balance sheet variables
of commercial banks
•Examine if the impact of monetary policy on balance
sheet variables of commercial banks is dependent on
regional groupings
Specific Objectives
7

8. METHODOLOGY
• The study employs dynamic panel data methodology, following González and Grosz (2014), Nana and
Samson (2014), Shrestha (2013) and Burgstaller (2010).
• 𝑦𝑖𝑡 = 𝑗=1
𝑝
𝛼𝑗 𝑦𝑖𝑡−𝑗 + 𝑥𝑖𝑡
′
𝛽 + 𝑤𝑖𝑡
′
𝛽 + 𝜇𝑖 + 𝑣𝑖𝑡 𝑖 = 1 … 𝑁, 𝑡 = 1 … 𝑇 (1)
• 𝜇𝑖 and 𝑣𝑖𝑡 are assumed to be independent for each 𝑖 and over all 𝑡
• Sources of persistence: autocorrelation caused by 𝑦𝑖𝑡−𝑗, and 𝜇𝑖 which illustrates heterogeneity.
• 𝑦𝑖𝑡−𝑗 will cause OLS estimator, FE estimator, or RE estimator, to be biased and inconsistent (Flannery
and Hankins, 2013)
• For many 𝑁 and few 𝑇 datasets, Arellano–Bond (1991) proposed GMM estimator constructed by first-
differencing to remove 𝜇𝑖 then using instruments to form moment conditions (Burgstaller, 2010).
• Arellano & Bover (1995), and Blundell & Bond (1998) however argued that the instruments used in the
estimation of difference GMM are weak as autoregressive process becomes too persistent or
𝜎 𝜇
𝜎 𝑣
→ ∞
• Hence Blundel and Bond (1998) system GMM estimator has been used to investigate effects of
monetary policy on: bank credit, liquid assets and deposits.
8

9. METHODOLOGY CONTI…
• The empirical model specifications that the one step system GMM has been applied to is:
∆𝑦𝑖𝑡= 𝛼∆𝑦𝑖𝑡−𝜌 + ∆𝑥𝑖𝑡
′
𝛽 + ∆𝑤𝑖𝑡
′
+ ∆𝑣𝑖𝑡 (2)
• The regressors are: real interest rates, capitalization, interaction term between real interest and
capitalization, international reserves, inflation rates, and GDP per capita growth.
• Tests for autocorrelation of orders one and two (Arellano/Bond 1991) have been used to ensure
efficiency of estimators.
• The validity of instruments has been evaluated using the Hansen (1982) J test which is robust to
heteroscedasticity but may be weakened with many instruments (Roodman 2009a,b).
Sources of Data
• The macro panel data used for the study covers the 2000 to 2014 period and includes 31 SSA countries
• The data used was mainly sourced from World Development Indicators (World Bank) and International
Financial Statistics (International Monetary Fund) databases.
9

10. RESULTS OF DESCRIPTIVE ANALYSIS 10
200
0
200
1
200
2
200
3
200
4
200
5
200
6
200
7
200
8
200
9
201
0
201
1
201
2
201
3
201
4
Bank Liquidity 26 20.8 21.7 24.7 25.7 27.3 28.7 28.5 27 26.4 27.1 22.9 25.8 23.3 26
Bank Credit 14.7 14.9 14.9 16.1 16.3 17.3 17.8 19.1 20.9 22 22.1 22.8 23.4 24.1 24.6
Bank Deposits 31.4 25.4 26.6 29.4 27.5 28.7 31.2 31.8 33.3 34.8 35.6 34.7 33.7 35.5 37.5
0
10
20
30
40
Average
Year
Balance Sheet Variables in SSA
Bank Liquidity Bank Credit Bank Deposits

11. RESULTS OF DESCRIPTIVE ANALYSIS 11
0
10
20
30
40
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
AverageBankCredit
Year
Average Bank Credit in SSA
ssa sadc ecowas eac cemac non grouped

12. DIAGNOSTIC TESTS
Bank Deposits
SSA SADC ECOWAS CEMAC EAC OTHERS
Arellano Bond Test -1.00 -1.02 -1.30 -1.32 -0.29 0.25
P-values 0.32 0.31 0.19 0.19 0.77 0.80
Hansen 15.02 0.00 0.00 0.00 0.00 0.00
P-values 1.00 1.00 1.00 1.00 1.00 1.00
Bank Credit
SSA SADC ECOWAS CEMAC EAC OTHERS
Arellano Bond Test -0.68 -0.54 -1.24 1.40 -0.64 1.02
P-values 0.49 0.59 0.22 0.16 0.52 0.31
Hansen 13.76 0.00 0.00 0.00 0.00 0.00
P-values 1.00 1.00 1.00 1.00 1.00 1.00
Liquid Assets
SSA SADC ECOWAS CEMAC EAC OTHERS
Arellano Bond Test 0.16 -0.76 -0.52 -0.21 1.43 0.23
P-values 0.87 0.45 0.61 0.84 0.15 0.82
Hansen 16.01 0.00 0.00 0.00 0.00 0.00
P-values 1.00 1.00 1.00 1.00 1.00 1.00
N 248.00 154.00 56.00 84.00 42.00 98.00
12

13. REGIONAL ANALYSIS FOR BANK DEPOSITS
13
Dependent Variable: Bank Deposits
SSA SADC ECOWAS CEMAC EAC OTHERS
Bank Credit−𝟏 1.000*** 0.976*** 0.762*** 0.949*** 0.960*** 0.984***
(0.008) (0.019) (0.036) (0.035) (0.016) (0.005)
𝐑𝐞𝐚𝐥 𝐈𝐧𝐭𝐞𝐫𝐞𝐬𝐭−𝟏 -0.034 0.076 -0.128*** -0.065** -0.037** -0.020
(0.038) (0.144) (0.025) (0.031) (0.017) (0.028)
𝐂𝐚𝐩𝐢𝐭𝐚𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧−𝟏 0.027 0.037 0.349 -0.264 -0.708* -0.058
(0.072) (0.099) (0.268) (0.376) (0.375) (0.064)
𝐂𝐚𝐩 ∗ 𝐑𝐞𝐚𝐥 𝐈𝐧𝐭.−𝟏 -0.004 -0.044 -0.082*** 0.017 0.004 0.002
(0.010) (0.039) (0.016) (0.019) (0.019) (0.003)
𝐅𝐨𝐫𝐞𝐱−𝟏 0.002 0.081 0.253** 0.064*** 0.181*** 0.100**
(0.025) (0.069) (0.127) (0.013) (0.039) (0.039)
𝐆𝐃𝐏−𝟏 0.104* -0.186 -0.061 0.049*** -0.197*** -0.001
(0.055) (0.322) (0.060) (0.018) (0.055) (0.121)
𝐈𝐧𝐟𝐥𝐚𝐭𝐢𝐨𝐧−𝟏 -0.088* -0.138* 0.066 0.007 -0.068 -0.012
(0.048) (0.071) (0.054) (0.059) (0.044) (0.010)
Wald Test 83873.98 33.21 13.29 68.42 13.68 21.43
P-value 0.00 0.00 0.00 0.00 0.00 0.00

14. REGIONAL ANALYSIS FOR BANK CREDIT
14
Dependent Variable: Bank Credit to the Private Sector
SSA SADC ECOWAS CEMAC EAC OTHERS
Bank Credit−𝟏 1.007*** 0.999*** 0.544*** 0.979*** 0.928*** 0.954***
(0.010) (0.010) (0.142) (0.030) (0.039) (0.025)
𝐑𝐞𝐚𝐥 𝐈𝐧𝐭𝐞𝐫𝐞𝐬𝐭−𝟏 -0.039* -0.111*** -0.004 -0.003 -0.110*** -0.072***
(0.021) (0.033) (0.028) (0.008) (0.037) (0.022)
𝐂𝐚𝐩𝐢𝐭𝐚𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧−𝟏 0.073 0.095 0.347** 0.330*** -0.989 0.046
(0.078) (0.096) (0.175) (0.123) (0.943) (0.031)
𝐂𝐚𝐩 ∗ 𝐑𝐞𝐚𝐥 𝐈𝐧𝐭.−𝟏 -0.012* -0.006** -0.023*** -0.017** -0.078*** -0.006***
(0.007) (0.003) (0.005) (0.007) (0.026) (0.001)
𝐅𝐨𝐫𝐞𝐱−𝟏 0.024* -0.004 0.324*** 0.038*** 0.385*** 0.156***
(0.014) (0.028) (0.124) (0.005) (0.079) (0.038)
𝐆𝐃𝐏−𝟏 0.101*** 0.105 -0.008 0.020 -0.348*** 0.147***
(0.037) (0.078) (0.040) (0.013) (0.048) (0.042)
𝐈𝐧𝐟𝐥𝐚𝐭𝐢𝐨𝐧−𝟏 -0.075*** -0.063 0.073 -0.069* -0.096*** -0.010***
(0.026) (0.063) (0.051) (0.039) (0.023) (0.003)
Ward Test 44930.95 51.65 27.05 11.50 22.22 35.45
P-value 0.00 0.00 0.01 0.04 0.02 0.00

15. RESULTS FOR REGIONAL GROUPINGS
15Dependent Variable: Bank Liquid Assets
SSA SADC ECOWAS CEMAC EAC OTHERS
Bank Credit−𝟏 0.822*** 0.895*** 0.599*** 0.488*** 0.746*** 0.789***
(0.049) (0.024) (0.093) (0.044) (0.040) (0.056)
𝐑𝐞𝐚𝐥 𝐈𝐧𝐭𝐞𝐫𝐞𝐬𝐭−𝟏 -0.050 -0.033 0.030 -0.299 0.256*** -0.057
(0.150) (0.071) (0.052) (0.252) (0.097) (0.050)
𝐂𝐚𝐩𝐢𝐭𝐚𝐥𝐢𝐳𝐚𝐭𝐢𝐨𝐧−𝟏 0.036 0.057 0.317** -1.362 3.338*** 0.325**
(0.384) (0.095) (0.143) (1.310) (0.453) (0.163)
𝐂𝐚𝐩 ∗ 𝐑𝐞𝐚𝐥 𝐈𝐧𝐭.−𝟏 -0.016 -0.024** -0.050*** -0.147** -0.134*** -0.002
(0.025) (0.010) (0.016) (0.072) (0.032) (0.007)
𝐅𝐨𝐫𝐞𝐱−𝟏 -0.007 0.055 0.011 0.430** -0.406*** -0.132
(0.098) (0.037) (0.120) (0.175) (0.126) (0.103)
𝐆𝐃𝐏−𝟏 -0.121 -0.219* -0.089 0.027 0.011 -0.298
(0.179) (0.131) (0.110) (0.120) (0.232) (0.240)
𝐈𝐧𝐟𝐥𝐚𝐭𝐢𝐨𝐧−𝟏 0.071 0.137*** 0.078 2.274** 0.187** -0.024***
(0.059) (0.047) (0.050) (0.883) (0.086) (0.006)
Wald Test 16020.98 16.60 15.54 19.32 14.33 16.64
P-value 0.00 0.01 0.03 0.01 0.04 0.01

16. CONCLUSION & POLICY IMPLICATIONS
• A lending channel in sub-Saharan Africa is a viable transmission mechanism
when the regional grouping are considered.
• The common monetary policy for CEMAC may be transmitted through a bank
lending channel to the economies of member states.
• Proposed common monetary policy for regions such as EAC and ECOWAS, once
formed, will notably be transmitted through a bank lending channel.
• A lending channel for SADC is not much certain because monetary policy stance
seem not to significantly affect bank deposits.
*******
THANK YOU
16