The Journal of Economic Asymmetries 29 (2024) e00355

Contents lists available at ScienceDirect

The Journal of Economic Asymmetries

journal homepage: www.elsevier.com/locate/jeca

Banking behaviour and political business cycle in Africa: The role

of independent regulatory policies of the central bank
Daniel Ofori-Sasu a, *, Elikplimi Komla Agbloyor a, e, Dennis Nsafoah b, Simplice
A. Asongu c, d
a
University of Ghana Business School, Ghana
b
Department of Economics and Finance, Niagara University, Lewiston, NY, 14109, USA
c
Department of Economic & Data Science, New Uzbekistan University, 54 Mustaqillik Avenue, Tashkent, 100007, Uzbekistan
d
School of Economics, University of Johannesburg, Johannesburg, South Africa
e
University of Stellenbosch Business School, South Africa

A R T I C L E I N F O A B S T R A C T

JEL classification: This study examines the effect of regulatory independence of the central bank in shaping the
D7 impact of electoral cycles on bank lending behaviour in Africa. It employs the dynamic system
D72 Generalized Method of Moments (SGMM) Two-Step estimator for a panel dataset of 54 African
G2
countries over the period, 2004–2022. The study found that banks lend substantially higher
G3
E3
during election years, and reduce lending patterns thereafter. The study shows that countries that
E5 enforce monetary policy autonomy of the central bank induce a negative impact on bank lending
E61 behaviour while those that apply strong macro-prudential independent action and central bank
G21 independence reduce lending in the long term. The study provides evidence to support that
L10 regulatory independence of the central bank dampens the positive effect of elections on bank
L51 lending around election years while they amplify the reductive effects on bank lending after
M21 election periods. There is a wake-up call for countries with weak independent central bank
P16
regulatory policy to strengthen their independent regulatory policy frameworks and political
P26
institutions. This will enable them better strategize to yield a desirable outcome of bank lending
Keywords:
to the real economy during election years.
Political economy
Political credit cycles
Electoral cycle
Central bank regulatory independence
Bank lending behaviour

1. Introduction

Recent debate in the literature highlights the procyclicality of bank lending behaviours during elections (Ghosh, 2022, 2020;
Koetter & Popov, 2021; Kumar, 2020; Englmaier & Stowasser, 2017). Drawing from experiences in several countries, the literature
shows that electoral events influence the lending behaviour of banks (see Ghosh, 2020; Englmaier & Stowasser, 2017; Carvalho, 2014;
Dinç, 2005; and many others). For instance, Englmaier and Stowasser (2017) provide evidence that banks that are controlled by

* Corresponding author.
E-mail addresses: [email protected] (D. Ofori-Sasu), [email protected] (D. Nsafoah), [email protected] (S.A. Asongu).

https://doi.org/10.1016/j.jeca.2024.e00355
Received 31 May 2023; Received in revised form 4 January 2024; Accepted 20 January 2024
Available online 22 February 2024
1703-4949/© 2024 Elsevier B.V. All rights reserved.
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

country-level politicians in Germany tend to adjust lending policies in response to local electoral cycle. This implies that as election
approaches, banks tend to overextend loans, often at the request of politicians (Ghosh, 2022). In particular, the response to bank
lending can be redistributed around election years conditioned on political business cycle and such targeted reallocation would often
be aimed to shift election outcomes in favour of the ruling party, or coalition parties in control of the central government (Bircan &
Saka, 2018). The political business cycles are cycles in macro-economic indicators such as inflation, unemployment and output, as well
as political institutional arrangements like the rule of law, government effectiveness, political stability and absence of violence, among
others, which are influenced by election cycles (Iddrisu & Ebo Turkson, 2020; Agbloyor, 2019). Thus, the political business cycles
influence the lending bahaviour of banks during electoral events.
The opportunistic view of political business cycle, according to Iddrisu and Ebo Turkson (2020, pp. 2), argues that “all govern­
ments, regardless of their political orientation, apply expansionary policies ahead of elections in order to increase their popularity and
brighten their chances in the re-election process.” This supports the mixed results in the literature concerning the effect of electoral
cycles on bank lending behaviour. While studies documented that political influence on state owned banks in emerging markets leads
to greater lending in election years (Englmaier & Stowasser, 2017; Carvalho, 2014; Dinç, 2005), in industrialized countries, there was
no discernible difference in the credit growth rates of public and private banks (Turkey et al., 2019; Dinç, 2005), and thus, banks hold
more capital through bank loan loss provisioning and lend less during election years across developed countries (Bircan & Saka, 2018,
2019; Ghosh, 2022; Ozili, 2019). Extant literature has ignored the African context and that the differences in the findings of earlier
studies could be attributed to the differences in political settings, business cycle and in particular differences in independent regulatory
reforms of institutions (Ghosh, 2022; Bircan & Saka, 2019, 2018; Ozili, 2019; Iddrisu & Ebo Turkson, 2020; Harris et al., 2018; Cohen
& Edwards, 2017; Dinç, 2005). In view of that, there is a good reason to expect regulatory policies to influence bank lending behaviour
across different electoral cycles. Although, the above discussion explains how the political economy influence bank lending behaviour
(see also, Dreher et al., 2019; Agarwal et al., 2016; Rodriguez & Santiso, 2008), little or no study has empirically examined the effect of
electoral cycles on bank lending behaviour from the African context. This study seeks to fill this gap.
It is commonly acknowledged that the regulatory body must be independent in order for regulatory decisions to be made and
enforcement measures to be carried out without unauthorized political interference or attempts to have a negative impact on price and
financial stability. Although there is a stronger analytical case for regulatory independence than there was in the past, not everyone
agrees that it is inherently desirable. In this paper, we examine the role that regulatory independence plays on the effect of electoral
cycles on bank lending behaviour. Our aim is to investigate whether the independent regulatory policies of the central bank help
African economies moderate (either reduce or enhance) the impact of electoral cycles on bank lending behaviour. Regulatory inde­
pendence of central bank, according to Thomson (2020), is a set of standards, policies and financial reforms that gives the monetary
authority the power to provide independent functions and regulatory decisions without political interference or the influence of
governments and political parties (see also, Müller, 2019; Viñals, 2013; Arnone et al. 2009). However, the concept of independent
regulatory policies and bank lending has attracted little attention in the literature. For instance, changes in political economy can cause
the central bank to act pro-cyclically or counter cyclically through the financial market around election periods (see, Keita & Turcu,
2022; Mpatswe et al., 2011), but what is not known is how independent regulatory policy affect the lending behaviours of bank.
Following the argument of the opportunistic model, stated above, it is important for policymakers and independent regulatory au­
thorities to understand the impact of independent regulatory policies of central bank in explaining the electoral cycle-bank lending
nexus.
Previous studies provide evidence that monetary policy influence banks’ pricing behaviour (Ciccarelli et al., 2015);
macro-prudential tools can stabilize credit growth (Ayyagari et al., 2017; BIS, 2017; Epure et al., 2017; Gambacorta & Murcia, 2017;
Jiménez et al., 2017) and in contrast to monetary policy, central banks are uniquely insulated from political cycles in macro-prudential
policy (Müller, 2019); bank’s lending patterns are affected by elections and business cycles (Ali et al., 2022); high bank loan prices in
election years increase access to finance compared to non-election years (Iddrisu & Ebo Turkson, 2020); and there is the existence of
the effect of political business cycles on economic growth and human development (Iddrisu & Mohammed, 2019; Mosley & Chir­
ipanhura, 2016). In addition, bank lending is constrained by monetary policy in emerging markets (Modugu & Juan Dempere, 2022;
Borio & Gambacorta, 2017; Altunbas et al., 2018; Amidu, 2006); macro-prudential policy efforts are successful in limiting excessive
credit booms (Altunbas et al., 2018; Cehajic & Kosak, 2021), and for reducing systemic risk (Akinci & Olmstead-Rumsey, 2018;
Meuleman & Vander Vennet, 2020); and banks with stringent macro-prudential policy have the incentive to shield their loan portfolio
(Fang et al., 2018; Jimenez et al., 2017). In line with government support, Azzimonti (2019) argues that several efforts of governments
in the implementation of macro-prudential policies and major reforms have played a vital role in mitigating excessive lending be­
haviours or risk-taking of banks and reducing the probability of financial crises, yet extant literature has been silent on testing the
independent regulation-lending nexus.
Despite the relevance of these issues, not much research has been undertaken on whether: (1) different electoral cycles impact bank
lending; (2) independent regulatory policies of central bank influence bank lending behaviour in election periods and in periods
without election; and (3) the independent regulatory policies of central bank influence the election-lending nexus. Africa is noted in its
paucity of funds required to grow the real sector of the economy and also, the combination of weak system stability, weak central bank
independence, independent regulatory reforms and misalignment among the fiscal, monetary policies and prudential regulations
(Gyeke-Dako et al., 2022; Strong, 2021; De Waal et al., 2018; Agoba et al., 2017; Arnone et al., 2009; Jeanneney, 2006), have created a
hug gap in the credit market (Amidu, 2006; Kanga, 2021). More so, Africa provides an interesting case study for this empirical
experiment because scholars and policymakers on the continent are now viewing independent regulatory policy framework of central
banks, as an important tool for stabilizing the banking system during electoral cycles (Nguyen et al., 2019). The continent offers a
conducive ground for the study of political business cycle since Africa is concurrently going through a prolonged process of economic

2
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

reform (Block, 2002).

Based on this, the study offers novel contribution to the literature by using the dynamic system Generalized Method of Moments
(SGMM) to examine how different electoral cycles affect bank lending in Africa. In addition, it employs different measures of inde­
pendent regulatory policies to provide a first time evidence of how these regulatory measures affect bank lending across different
electoral regimes in Africa. It also makes significant contributions to the literature by examining the role of independent regulatory
policies of the central bank in moderating the relationship between electoral cycles and bank lending behaviour in Africa.
The rest of the study is organized into five sections. Overview and Literature review of related studies are contained in section 2 and
section 3 respectively, section 4 discusses the data and methodology. The empirical results are contained in section 5 and section 6
concludes the study.

2. Literature review: theories, empirics and hypothesis development

Theories of political lending cycles predict that governments intervene in the banking business and that the engagements of
governments in the banking business affect the banking sector. Banks that are under government control are politically motivated and
they are constrained with capital during electoral periods (Gerschenkron, 2015). This is because governments use loans by state-owned
banks as a strategic tool for re-election (Bircan & Saka, 2019). In particular, bank credit policy can be significantly adjusted around
election years. This adjustment in credit policy during election periods can have real effects on the economy. For instance, the banking
sector provide banking services through lending channels and that the public banks play an important role in times of crisis by
providing loans and by ensuring market liquidity (Brei & Schclarek, 2013; Carvalho, 2014; Chen et al., 2014). The study is perceived
through the lens of economic institutional theory, which focuses on the roles of social, political and economic systems in which
companies operate and gain their legitimacy (Shrum et al., 2001). Thus, politicians can compel banks to make loans to politically
connected companies under favourable terms, such as interest rates and long maturities for the loans (Meriläinen, 2016; Micco &
Panizza, 2006; Shleifer & Vishny, 1994).
Government involvement in the banking industry has an impact on bank lending behaviour. The literature provides evidence that
countries in underdeveloped regions increase the rate of loan growth of public banks during elections, whereas in developed nations, it
is not different from the loan growth rate of private banks (Ali et al., 2022; Shleifer & Vishny, 1994). This confirms that lending by
public banks is less pro-cyclical than lending by private banks in nations with solid governance (Bertay et al., 2015). Ghosh (2022)
investigates the impact of elections on bank provisioning for a longitudinal dataset of India, and found that banks reduce provisions
around elections and it is profound in state-owned banks. In addition, Gerschenkron (2015) indicated that government instruct banks
to supply capital to individuals who need access to funding. Dinç (2005), used a dataset of 36 emerging markets and developed
economies, over the period 1994–2000, to study the behaviour of lending during elections. He revealed that, during elections, public
bank lending increases in developing countries. However, none of the studies in the literature looked at how electoral cycles (dynamics
of election events) affect the lending behaviour of banks in Africa.
This study contributes to the literature by empirically testing the hypothesis stated below:
H1. : Electoral cycles are important in determining the levels of bank lending behaviour in different political business regimes
The focus of regulations on the banking behaviours across business cycles has a solid foundation in building a strong social and
political economy. On the regulatory independence, there is generally low central bank independence in African countries; thus, the
central banks are not free from interference from the incumbent government in their conduct of monetary policy, macro-prudential
and governance policy framework. In view of that, the central bank is always influenced to embark on expansionary policies,
particularly in election years (Iddrisu & Ebo Turkson, 2020). On one hand, monetary and macro-prudential instruments of the central
bank are seen to influence banking behaviours (see, Modugu & Juan Dempere, 2022; Hodula & Ngo, 2021; Cehajic & Kosak, 2021;
Abuka et al., 2019; Ayyagari et al., 2017). For instance, Modugu and Juan Dempere (2022) examine the impact of monetary policy
instruments on bank lending in the emerging economies of Sub-Sahara Africa, using the dynamic system generalized method of
moments (GMM) for 80 banks in 20 Sub-Sahara Africa over the 2010–2019 period. They found interesting results by showing that
expansionary monetary policy (i.e., loosening of the policy rate and increasing of money supply) propels bank lending while monetary
contractions (tightening of monetary policy rates and reducing money supply) by the central bank leads to credit contraction. In
addition, monetary policy is found to be a weak bank lending channel in developing countries (Abuka et al., 2019). They provided new
evidence that contractionary monetary policy reduces credit supply, leading to greater rejection of loan applications and the tightening
of lending rates and volumes. In a study by Cehajic and Kosak (2021), they analyze the effects of macro-prudential measures on bank
lending in the European Union. They employed 3434 European banks with 18,616 observations covering the period between 2000 and
2017. They found that macro-prudential instruments are used effectively by regulatory authorities for modulating credit activities of
banks across the business cycles. They provide evidence to support that in periods of loosening cycles, macro-prudential measures are
positively associated with bank lending. However, the impact is weak during periods of tightening actions, where the measures of
macro-prudential policies are found to have a downward effect on bank lending. Evidence from non-bank credit intermediation in 23
European Union countries indicates that macro-prudential actions affect shadow lending (see Hodula & Ngo, 2021). Hodula and Ngo
(2021) applied an instrumental variable (IV) estimation framework to demonstrate that the tightening of macro-prudential policy
leads to an increase in bank lending. Additionally, the impact is more binding in a low-capitalized banking system, leading to credit
restructuring and reallocation from banks to the non-banking sector. Ayyagari et al. (2017) combined data on 1.3 million firms from
2002 to 2011 operating in 59 countries that have undergone some changes in macro-prudential regulations over the period. They
found evidence to support that macro-prudential policies are important in lowering credit growth.

3
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

On the other hand, monetary autonomy of the central bank affects the lending channels of the banks. For instance, monetary
autonomy under fixed exchange rate regime affect the lending behaviour of banks (Rey, 2016; Farhi & Werning, 2014; Klein &
Shambaugh, 2015); and by the recent work from Muller (2020) showing that macro-prudential regulation is influenced by electoral
cycles. The monetary policy is conducted by the Central Bank of West Africa States, which is empowered to take any measures con­
cerning instruments and rules related to the credit policy applicable to credit institutions. In the West African Economic and Monetary
Union (WAEMU), capital mobility is therefore restricted and can lead to monetary policy independence. There has been extensive
debate about the political business cycles and their impact on several other outcome variables. Interestingly, the emerging literature
has significantly illustrated the existence of political business cycles with few relating to economic growth and development (Funa­
shima, 2016). In addition, central banks in advanced economies can sacrifice some political independence without undermining the
operational independence to their monetary policy and financial stability functions. However, what is missing in the literature, in
particular Africa, is whether the independence of regulatory policies of central bank affect bank lending behaviours. This study
provides insights into the response of bank lending behaviour to regulatory independence of central banks in Africa. We formulate the
following hypothesis:
H2. Regulatory policy independence of the central bank has a significant impact on bank lending behaviours across electoral regimes
From the theoretical and empirical reviews, it is evident that regulatory independence of the central bank plays a significant role in
election-lending nexus, it may either reduce or magnify the impacts. However, empirical studies to this effect are nonexistent in Africa.
Interestingly, following the literature on central bank’s regulatory independence and institutional arrangements that may differ across
regions (Gabriel et al., 2022; Jones, 2022; Klüh & Urban, 2022; Martinez-Resano, 2004; Romelli, 2022; Satragno, 2022, p. 228), the
individual impact of monetary policy (Borio & Gambacorta, 2017; Kakes & Sturm, 2002; Mishra et al., 2014; Modugu & Juan Dempere,
2022; Mwankemwa & Mlamka, 2022; Yun & Cho, 2022), macro-prudential (Altunbas et al., 2018; Auer et al., 2022; Czaplicki, 2022;
Hodula & Ngo, 2021) and central bank policy independence (Abor et al., 2022; Agoba et al., 2020; Doumpos et al., 2015) on bank
lending behaviour can vary across different institutional framework. However, the literature is silent on this. In this study, we attempt
to present first time evidence on how the independent central bank regulatory policy shapes the impact of electoral cycle on bank
lending behaviour. Thus, we test the hypothesis that:
H3. : Regulatory policy independence is important in shaping the effect of electoral cycles on bank lending behaviours

3. Data and methodology

The study employs a panel dataset of 54 African economies covering the period, 2004–2022. The sample includes countries that
have experienced presidential elections and have undergone structural reforms at any time during the sample period. The panel
approach enables us to account for continuously evolving country-specific differences in technology, institutional and economic
factors.
We utilize the baseline model, which is expressed as:
Bank lending behaviour = f (Electoral Cycles, Regulatory Independence, Control variables) (1)
Following Dinç (2005, pp.472), we address potential endogeneity using the dynamic system Generalized Method of Moments
(SGMM) approach. We use the strength of the independent regulatory policy of the central bank as well as their lags and leads in
pursuing its goals to instrument for differences in the use of regulatory policy measures across countries. The assumption underlying
the selection of instruments is supported by several research works (e.g., Bodenstein et al., 2019; Carrillo et al., 2021; Paoli & Paustian,
2017; Wintoki et al., 2012). Thus, we employ the dynamic system Generalized Method of Moments (SGMM) Two-Step estimation
technique.

3.1. Model specification

We begin our empirical analysis by considering the number of lags of bank lending which are adequate for capturing the dynamic
completeness of our benchmark model. In this regard, previous literature recommends the use of two lags for capturing the influence of
past indicators on current data (see for example, Wintoki et al., 2012; Gschwandtner, 2005). First, the study seeks to examine the
impacts of electoral cycle and regulatory independence of the central bank on bank lending behaviour. Second, it examines the
interactive effect of electoral cycle and regulatory independent on bank lending.

3.1.1. Impacts of electoral cycle and regulatory independence on bank lending behaviour
From the baseline equation, we estimate the independent effect of electoral cycle and regulatory independence of the central bank
on bank lending behaviour by following the works of Koetter and Popov (2021) and Iddrisu and Ebo Turkson (2020). This allows us to
specify the dynamic SGMM equation below:

4
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

p
∑ ∑
2 ∑
2
Bank lending behaviourjt = σg Bank lending behaviourj,t− g + α0 Electoral cyclejt + αl − Electoral cycle j,t− l + αl + Electoral cycle j,t+l
g=1 l=1 l=1

∑
3 ∑
3 ∑
3 ∑
N
+ ƛi Regulatory independencejt + ƛi − Regulatory independencej,t− i + ƛi + Regulatory independencej,t+i + βk Xjt + γ j + μt + εjt
i=1 i=1 i=1 k=1

(2)

where subscript j denotes cross sectional dimension (country specifics), j = 1, …, M; t denotes the time series dimension (time).
In equation (2), t = 1, …, T; and t − g, g = 1, …, 2, denote the lag dimensions of bank lending behaviour; t-l, l = 0, …, 2, denote the
lag dimensions of electoral cycles; and t +l, l = 1, …, 2, denote the lead dimensions of electoral cycles ; σ g : g = 1, ...p, represent the
regression coefficients of the lags of the dependent variable; α0 is the regression coefficient of elections in the current year t; αl − : l =
1, ..., 2, represent the regression coefficients of a vector of the lags of electoral cycle; αl + : l = 1, ..., 2, represent the regression co­
efficients of a vector of the leads of electoral cycle; ƛi : i = 1, ..., 3, denote the regression coefficients of three individual indicators
capturing the central bank regulatory independence in the current year t; ƛi − : i = 1, ..., 3, represent the regression coefficients of the
lags of the three indicators capturing regulatory independence of the central bank, ƛi + : i = 1, ..., 3 represent the regression coefficients
of the leads of the three indicators capturing regulatory independence of the central bank; t-1 and t +1 represent the lag and lead of
regulatory independence respectively; βk : k = 1,...,N, are regression parameters for a vector X (measuring a set of control variables) to
be estimated; γj is the country fixed effect; and μt is the time fixed effect t; and εjt is idiosyncratic error term, which controls for unit-
specific residual in the model for the jth country at period t.

3.1.1.1. Measurements. In equation (2), the impact of electoral cycle and regulatory independence on bank lending are estimated
independently before estimating their interactive effects.

3.1.1.2. Bank lending behaviour. The dependent variable in equation (2), bank lending behaviour, is measured using the percentage of
aggregate bank credit to gross domestic product (GDP) in a given country. It indicates the average level of participation of the banking
sector in the real economy, as used in the literature by (Amidu, 2006; Borio & Gambacorta, 2017; Abuka et al., 2019; Modugu & Juan
Dempere, 2022). This includes the volume of loans to the state government, corporates, businesses and households. Data on bank
credit to GDP was obtained from the World Bank Global Financial Development Database. An increasing level of bank credit to GDP
shows greater lending behaviour of the banking sector.

3.1.1.3. Electoral cycles. Elections are events which motivate the politicians to use government’s resources to increase their chances of
election. The electoral cycle in our model is the period around a country’s defined election year. Following Koetter and Popov (2021),
Iddrisu and Ebo Turkson (2020), Agbloyor, (2019), we construct electoral cycle variables as a dummy and the variables were obtained
mainly by searching online to find out when elections were held. Using the dummy identifying elections, we decompose electoral cycle
into three (3), and this includes: (1) election period (denoted as Electoral cycle t ; constructed as a dummy, with a value of 1 in election
years and 0 otherwise); (2) pre-election denoted as Electoral cycle t− l ; constructed as a dummy, with a value equal to 1 in years before
presidential elections, and 0 otherwise); and (3) post-election (denoted as Electoral cycle t+l ; constructed as a dummy, with a value
equal to 1 in years after presidential election, and 0 otherwise. A post-election period captures a national election after which either a
new party comes into power or there is a continuity of incumbent government. In constructing the post-election dummy, we differ­
entiate between elections with a change in government from all other elections. This allows us to rule out the tendency that changes in
lending patterns observed after power-changing elections are strictly driven by elections themselves, regardless of the outcome. In
robustness tests, we use one pre-and two-post election observations for elections in our dataset typically take place at 4-year or 5-year
intervals and only exceptionally take place at 2-year or 3-year periods. In equation (2), we include real GDP per capita and individual
measures of political institutional variables as controls to observe the behaviour of the impact of electoral cycles on bank lending
behaviour. Based on this, we expect varying results between electoral cycle and bank lending behaviour. For instance, we expect banks
to increase their lending before presidential election as supported by Fungáčová et al., (2020). However, we expect banks’ to either
increase or lower their lending capacity during and after election periods due to possible risks, political, government and public in­
terests (see for example, Koetter & Popov, 2021).

3.1.1.4. Regulatory independence. The Organisation for Economic Co-operation and Development (OECD) describes regulatory in­
dependence as the “protection from attempts to exercise undue control and influence from government and stakeholders external to
the regulator and those who seek to inappropriately influence regulatory decision-making from within (OECD, 2014, 2017, pp.5).”
Regulatory independence allows the central bank to independently focus on financial and price stability goals through monetary and
macro-prudential instruments as well as governance roles, supervision and responsibilities (Balls et al., 2018; Cukierman, 2008, 2013;
Kohn, 2015). The study draws up an ideal framework for measuring regulatory independence of the central bank. Thus, we decompose
regulatory independence of the central bank into three individual regulatory measures: (1) Monetary policy independence; (2)
Macro-prudential independence; and (3) Central Bank independence. In equation (2), we simultaneously introduce the individual
variables of the central bank regulatory independence into the model in order to examine their independent effect on bank lending.
Monetary Policy independence is the policy actions that give power to the central bank to control its own monetary policy
instrument for domestic purposes independent of external monetary influences. Example is the independent control of the policy rate

5
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

or interest rate payable on short-term borrowings. Monetary policy independence is not easy to define and measure, but one of the
widely used measures is the extent of deviation of the domestic interest rate from the base rate. We employ the monetary independence
index defined by Aizenman et al. (2020), as the reciprocal of the annual correlation between the monthly short-term interest rates of
the home country and the base country. We obtained data on monetary policy independence from IMF’s International Financial
Statistics (IFS). Monetary policy independence index ranges between 0 and 1 with higher values of index denoting lower correlation of
interest rates and thus greater monetary policy independence. We expect monetary policy independence to reduce the level of bank
lending behaviour. This supports empirical works which directly test and lend support to the notion that countries with more inde­
pendent monetary instrument tend to deliver better inflation outcomes, leading to contractionary policy, and thus, lowering bank
lending.
Macro-prudential independence is the approach to prudential regulations that aim to offer regulatory authorities the sole
mandate to supervise and mitigate possible risk to the financial system as a whole. Edge and Liang (2019) focuses on the degree of
macro-prudential independent policy by placing a relatively low weight on the ability of a country’s policy institutions to take action
and placing a high weight on political economy considerations in developing a country’s financial stability governance structures (see
also, Sever & Yücel, 2022; Masciandaro & Romelli, 2018). Contrary to their measure, we employ the Alam et al. (2019) measure of
macro-prudential independent policy. An independent macro-prudential authority is given the power to make policy decisions and
enable it to perform effectively (Krishnamurti & Lee, 2014). According to Buch et al. (2018), a structured policy process can be a key
element in ensuring that prudential policy decisions are based on independent assessments, on transparent decisions, and that
decision-makers are accountable to the public. Given that price stability and financial stability policies are closely interlinked, dele­
gating macro-prudential authority to a government agency other than the central bank may threaten the bank’s independence over its
objective of maintaining stability in the financial economy (see, Duff, 2014). There is no standard definition and measure of
macro-prudential independence, we rely on Alam et al. (2019) databases of macro-prudential index as the indicators of
macro-prudential instruments are often determined independently by the central bank. It is constructed as an index of dummies. This is
the policy change indicator for the instrument which records tightening actions (+1) (i.e. strengthening), loosening actions (− 1) (i.e.
relaxing), and no changes (0), and it is cumulated over the past four quarters to account for potential lagged effects. Data on
macro-prudential policy is an aggregate (composite) index of 17 indicators of macro-prudential action (countercyclical capital buffer,
requirements for banks to maintain a capital conservation buffer, capital requirements, limits on leverage of banks, loan loss provision
requirements, limits on foreign currency, limits to the loan-to-value ratios, debt service-to-income ratio, minimum requirements for
liquidity coverage ratios, limits to the loan-deposit ratio, limits to net or gross open foreign exchange positions, reserve requirements,
loan restrictions, risk measures, taxes and levies applied to specified transactions, and macro-prudential measures not captured in the
above categories). These are sum of all the dummies of the policy actions recorded in the databases and takes values from − 1 to 1, with
higher values of the index indicating strong macro-prudential independence. Data was obtained from the iMaPP database constructed
by Alam et al. (2019), integrating information from major existing data basses (the Global Macro-prudential policy instruments and
IMF annual macro-prudential policy survey), national sources (Lim et al., 2011; 2013; Alam et al., 2019). We expect a negative effect of
macro-prudential independence on bank lending. Given that macro-prudential tools are structured with the objective to increase the
resilience of the financial sector, increasing the level of central bank regulatory independence tends to control banks’ capital reserves
and induce greater restriction on their risky lending behaviours. This is consistent with the works of Behncke (2022), Abuka et al.
(2019) and Hussain and Bashir (2019).
Central bank independence is an index that measures the ability of the central bank to formulate independent policies, as
employed in the work of Agoba et al. (2020). In general, it is a policy that controls monetary policy tools and limits the government’s
influence on the management of monetary policy by the central bank. It is a de jure measure of central bank independence based on a
weighted aggregation of 16 legal indicators using the criteria and weights of the Cukierman, Webb and Neyapti indexes (CWN)
(Garriga, 2020). The index varies between 0 and 1 (i.e., 0 and 100 %), with higher values indicating a greater degree of central bank
independence or a more stringent independent central bank. The study expects that central bank independence should have a negative
effect on the lending behaviour of banks. The independent role of central banks enables them to monitor the opportunistic behaviour of
managers, control excessive risk-taking and achieve optimal returns. This requires banks to reduce output (i.e. loans or lending) while
raising prices (interest rates) to yield more returns, thus inducing a negative impact on bank lending behaviour (see, Behncke, 2022;
Abuka et al. 2019; Hussain & Bashir, 2019).
In addition, it might be argued that central bank independence may go against democratic ideals to have unelected central bankers
make significant decisions about economic policy. In a different political context, the value of independent central banks may be
questioned on the grounds that they may not actually deliver superior monetary policy outcomes – and therefore affect bank lending
outcomes. For this reason, we expect that the impact of each regulatory independence on bank lending should differ across different
electoral regimes (periods of election and periods without election).
In addition, for robustness checks, we assume that fluctuations of the political economy can cause the central bank to independently
act pro-cyclically or counter cyclically ((see, Keita & Turcu, 2022; Mpatswe et al., 2011), leading to changes in bank lending behaviour.
For this reason, we introduce the lag and lead terms of individual regulatory independence into equation (2). We expect the dynamics
of the individual regulatory independence, based on its lag and lead to have a significant impact on bank lending.
In equation (2), X is the set of control variables. All control variables are described in Appendix A.

3.1.2. Interactive effects

In this section, we seek to test whether regulatory independence of the central bank amplifies the relationship between electoral
cycle and bank lending behaviour. Because electoral cycle and regulatory independence may have independent impact on bank

6
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

lending behaviour, we also estimate an equation to capture the interactive effect between electoral cycle and regulatory independence.
This is specified as:
∑
2 ∑
2
Bank lending behaviourjt = Φ1 Bank lending behaviourj,t− 1 + β0 Electoral cyclejt + βl − Electoral cycle j,t− l + βl + Electoral cycle j,t+l
l=1 l=1
∑
3 p
∑ ( )
+ ƥi Regulatory independence jt + δq Electoral cycle jt *Regulatory independencejt
i=1 q=1
p
∑ ( )
+ δq − Electoral cycle j,t− l *Regulatory independencejt
q=1
p
∑ ( ) ∑N
+ δq + Electoral cycle j,t+l *Regulatory independencejt + αk Cjt + σ j + θt + μjt
q=1 k=1

(3)

where δq , δq − and δq + : q = 1, ..., p denote the coefficients of the interaction terms between the respective electoral cycle variables
(contemporaneous, lags and leads) and the individual central bank regulatory independence variables; Φ1 represents the coefficient of
the first lag of the dependent variable ; β0 is the regression coefficient of elections in current year; βl − and βl + : l = 1, ..., 2, represent the
regression coefficients of a vector of the respective lags and leads of electoral cycle variables; ƥi : l = 1, ..., 3, represent the regression
coefficients of a vector of three indicators capturing regulatory independence of the central bank; αk , k = 1, …, N are the coefficients of
the control variables (for vector C); σ j is the individual country effects; and θt is the time fixed effects and μjt is the composite error term.
To examine the relevance of individual regulatory independence variables in explaining the impact of electoral cycle on bank
lending, net impacts are calculated for equation (3), in accordance with contemporary studies on interactive regressions (Agoba et al.,
2020; Asongu & Odhiambo, 2021). The computation of the net effect entails the sum of the unconditional impacts of electoral cycle
and the conditional or marginal effects of electoral cycles.
Thus, we observe the net impact of the electoral cycle on bank lending when interacted with the individual regulatory indepen­
dence variables.
From equation (3), the net effects are expressed below:
∂Bank lending behviourj,t
Net Effect = = β0 + δq Regulatory independencejt (4)
∂Electoral cyclejt

where β0 is the regression coefficient of the year of elections; and δq are the coefficients of the interaction terms.
For robustness test, we examine the net effects of electoral cycle and regulatory independence of the central bank on bank lending
in a country with strong political institution and those in weak political institution. We do this by splitting the samples into two, that is,
we decompose political institution variable into strong political institution (equal to 1 if the average political institution of a country is
equal to or greater than the overall average in the sample) and weak (equal to 1 if the average political institution of a country is strictly
less than the overall average in the sample). We observe the net effects across the two samples.

3.2. Estimation technique

To enhance reliability, efficiency and accuracy of the result, the study employs a number of techniques. We begin our empirical
analysis by considering the number of lags of the variables which are adequate for capturing the dynamic completeness of our model.
We test the optimal lag/lead length using the Akaike or Schwarz information criterion (AIC or BIC). The AIC helps to select the optimal
model that gives the lowest values of the criteria, while the autocorrelation function (ACF) and Ljung-Box tests help us to quantitatively
test for autocorrelation at multiple lags/leads jointly. We use a year lag for our variables in the model, based on the selection criterion,
because introducing more lags might lead to the likelihood of producing inefficient parameter estimates or the standard errors of the
regression coefficients, multicollinearity among the regressors, serial correlation in the error terms as well as misspecification errors.
We treat all variables except the year dummies as endogenous to make room for the instruments of all those explanatory variables
which are not strictly endogenous. This allows us the use of an additional lag at year 1 of all such variables as an instrument. It has been
argued that the selection of instruments is based on unrealistic assumptions of data, leading to the use of instruments that are not
totally exogenous (see, Aggarwal et al., 2009). We employ the dynamic System Generalized Method of Moments (SGMM) Two-Step
estimator with small sample size adjustments, forward orthogonal deviations and robust standard errors. This allows for the use of
past values of the electoral cycle as instruments and thus, improving efficiency and reduces finite sample bias (see Arellano & Bover,
1995). The GMM resolves issues of unobserved heterogeneity that may arise between countries and endogeneity that may exist from
bi-causality and mismeasurements. To correct endogeneity, the System GMM technique introduces more instruments for the lagged
dependent variable and any other endogenous variable to drastically enhance efficiency, and it transforms the instruments to make
them uncorrelated (exogenous) with fixed effects. The use of system GMM helps to generate its own instruments from the data. We
report Hansen and Sargan tests. Hansen J test is used to test the validity of Instruments: tests the null hypothesis of overall validity of
instruments; failure to reject these null hypotheses gives support to the choice of the instruments. The Hansen test is distributed as
chi-square under the null shows that the instruments are valid. The validity of the test shows that the null hypothesis that “the

7
 Table 1

D. Ofori-Sasu et al.
Impacts of Electoral cycles on Bank Lending Behaviour.
VARIABLES Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Model 8

Bank lendingt-1 − 1.272*** − 2.809*** − 2.812*** − 2.769*** − 2.811*** − 2.798*** − 2.777*** − 2.801***
(0.199) (0.654) (0.638) (0.676) (0.638) (0.626) (0.677) (0.599)
Electoral cycle t-1 0.0263*** 0.0274*** 0.0249*** 0.0264*** 0.0236*** 0.0260*** 0.0311***
(0.00901) (0.00903) (0.00906) (0.00904) (0.00905) (0.00906) (0.00902)
Electoral cycle 0.235*** 0.0516*** 0.0505*** 0.0534*** 0.0518*** 0.0544*** 0.0527*** 0.0480***
(0.0854) (0.00976) (0.00978) (0.00981) (0.00979) (0.00990) (0.00983) (0.00973)
Electoral cycle t+1 ¡0.00107*** ¡0.00122*** ¡0.00122*** ¡0.00103*** ¡0.00108*** ¡0.00109*** ¡0.00106***
(0.000257) (0.000303) (0.000303) (0.000264) (0.000264) (0.000255) (0.000255)
Electoral cycle t+2 ¡0.0585*** ¡0.0580*** ¡0.0598*** ¡0.0586*** ¡0.0599*** 0.0587*** 0.0540***
(0.0110) (0.0110) (0.0111) (0.0111) (0.0110) (0.0111) (0.0110)
Banking crisis − 0.0398** − 0.0426*** − 0.0383** − 0.0327** − 0.0434*** − 0.0310* − 0.0341**
(0.0164) (0.0164) (0.0162) (0.0160) (0.0166) (0.0160) (0.0169)
Bank concentration − 6.20e-05*** − 6.56e-05*** − 6.32e-05*** − 6.85e-05*** − 5.65e-05*** − 6.91e-05*** − 4.62e-05**
(1.70e-05) (1.66e-05) (1.70e-05) (1.65e-05) (1.67e-05) (1.65e-05) (1.80e-05)
Credit risk − 0.000270 − 0.000242 − 0.000213 − 0.000321* − 0.000152 − 0.000261 − 0.000360*
(0.000185) (0.000185) (0.000186) (0.000186) (0.000186) (0.000186) (0.000184)
Foreign entry 0.0356 0.0399* 0.0445* 0.0459* 0.0293 0.0564** 0.0208
(0.0240) (0.0235) (0.0234) (0.0244) (0.0235) (0.0235) (0.0248)
inflation − 0.132*** − 0.0260*** − 0.0260*** − 0.118*** − 0.0718*** − 0.125*** − 0.0766***
(0.0117) (0.00808) (0.00808) (0.0111) (0.0101) (0.0113) (0.0103)
Real GDP per capita 0.0158*** 0.0159*** 0.0156*** 0.0155*** 0.0159*** 0.0155*** 0.0161***
(0.00157) (0.00154) (0.00158) (0.00158) (0.00159) (0.00157) (0.00158)
Political institutions 0.0199***
(0.00593)
8

Control of corruption 0.0235***

(0.00632)
G-Effectiveness 0.0157**
(0.00623)
Pol. Stability-Violence 0.00675*
(0.00398)
Regulatory quality 0.0270***
(0.00558)
Rule of law 0.00351

The Journal of Economic Asymmetries 29 (2024) e00355

(0.00570)
Voice and accountability 0.0264***
(0.00533)
Time Fixed Effect Yes Yes Yes Yes Yes Yes Yes Yes
Constant 14.15 0.496*** 0.495*** 0.482*** 0.479*** 0.500*** 0.465*** 0.520***
(10.86) (0.0260) (0.0251) (0.0251) (0.0266) (0.0253) (0.0251) (0.0271)

Observations 775 775 775 775 775 775 775 775

Number of Group 37 46 45 48 48 48 46 45
Instrument 22 27 26 23 23 23 27 26
AR (1) z (p value) − 1.61 (0.09) − 3.29 (0.001) − 3.33 (0.001) − 3.38 (0.001) − 3.50 (0.001) − 3.27 (0.001) − 3.29 (0.001) − 3.33 (0.001)
AR (2) z (p value) 3.4 (0.454) 0.65 (0.517) 0.68 (0.509) 0.63 (0.499) 0.65 (0.517) 0.61 (0.491) 0.65 (0.517) 0.68 (0.509)
Sargan Test (p-value) 15.93 (0.145) 14.47 (0.208) 15.46 (0.162) 15.90 (0.145) 20.67 (0.393) 22.30 (0.542) 14.47 (0.208) 18.41 (0.763)
Hansen Test 5.20 (0.270) 21.13 (0.142) 7.706 (0.565) 5.15 (0.272) 5.07 (0.397) 5.20 (0.270) 9.067 (0.431) 12.12 (0.335)
Chi2 (p) value)

Fisher 960.99*** 1985.51*** 2160.99*** 2808.65*** 2156.63*** 2160.99*** 1985.51*** 2160.99***

Table 1 shows the effect of electoral cycles on bank lending behaviour. All variables are described in Appendix A.
 D. Ofori-Sasu et al.
Table 2
Impacts of central bank regulatory independence on bank lending across different electoral regimes.
Monetary policy independence Macro-prudential independence Central bank Independence

Full sample Periods without Election Full sample Periods without Election Full sample Periods without Election
election periods election periods election periods

VARIABLES Model 9 Model 10 Model 11 Model 12 Model 13 Model 14 Model 15 Model 16 Model 17

Bank lending t-1 − 0.765*** − 1.229*** 0.446** − 0.654*** − 0.770** 0.698** − 0.679** − 0.845*** 0.512*
(0.198) (0.401) (0.165) (0.136) (0.365) (0.281) (0.271) (0.143) (0.298)
Bank lending t-2 − 0.0714*** − 0.0650*** 0.101*** − 0.0756*** − 0.0532*** 0.0979*** − 0.0808*** − 0.0165*** 0.00747***
(0.0149) (0.0133) (0.0145) (0.0139) (0.0136) (0.0135) (0.0126) (0.00416) (0.00186)
Monetary policy independence ¡0.00792** ¡0.00251* ¡0.00280**
t-1
(0.00384) (0.00145) (0.00124)
Monetary policy independence ¡0.0266*** ¡0.0273*** ¡0.134***
(0.00919) (0.00996) (0.0159)
Monetary policy independence ¡0.0339*** ¡0.0303** ¡0.117**
t+1
(0.0108) (0.0121) (0.0561)
Macro-prudential 0.0160*** 0.00873*** ¡0.0958*
independence t-1
(0.00490) (0.00244) (0.0491)
Macro-prudential 0.0319*** 0.0368*** ¡0.112**
9

independence
(0.00689) (0.00703) (0.0556)
Macro-prudential ¡0.0985*** 0.127*** 0.0104***
independencet+1
(0.0137) (0.0141) (0.00240)
Central bank Independence t-1 0.0132*** 0.00654*** 0.146***
(0.00259) (0.00242) (0.0145)
Central bank Independence 1.014*** 0.0886*** 0.00718***
(0.359) (0.0157) (0.00266)
Central bank Independencet+1 0.00328** 0.00196**

The Journal of Economic Asymmetries 29 (2024) e00355

¡0.892**
(0.363) (0.00162) (0.000826)
Election event − 0.168*** − 0.149*** − 0.150***
(0.0163) (0.0160) (0.0177)
Banking crisis − 0.0398** − 0.0426*** − 0.0383** − 0.0327** − 0.0434*** − 0.0310* − 0.0383** − 0.0327** − 0.0434***
(0.0164) (0.0164) (0.0162) (0.0160) (0.0166) (0.0160) (0.0162) (0.0160) (0.0166)
Bank concentration − 0.0158*** − 0.0159*** − 0.0156*** − 0.0155*** − 0.0159*** − 0.0155*** − 0.0159*** − 0.0156*** − 0.0155***
(0.00157) (0.00154) (0.00158) (0.00158) (0.00159) (0.00157) (0.00154) (0.00158) (0.00158)
Credit risk − 0.000321* − 0.00489* − 0.00561** − 0.0106* − 0.000360* 0.00580** − 0.0106* − 0.00793*** − 0.00580**
(0.000186) (0.00281) (0.00268) (0.00523) (0.000184) (0.00222) (0.00523) (0.00240) (0.00222)
Foreign entry 0.189*** 0.488*** 0.0247 0.212 0.364** 1.371* 0.148* 0.490*** − 0.0784
(0.0583) (0.0960) (0.0548) (0.135) (0.158) (0.708) (0.0766) (0.126) (0.115)
Inflation − 0.00424*** − 0.00594*** − 0.00242* − 0.000456 − 0.000784 − 0.00657*** − 0.00262** − 0.00430*** − 0.00509***
(continued on next page)
 D. Ofori-Sasu et al.
Table 2 (continued )
Monetary policy independence Macro-prudential independence Central bank Independence

Full sample Periods without Election Full sample Periods without Election Full sample Periods without Election
election periods election periods election periods

VARIABLES Model 9 Model 10 Model 11 Model 12 Model 13 Model 14 Model 15 Model 16 Model 17

(0.00114) (0.00153) (0.00143) (0.00198) (0.00230) (0.00192) (0.00116) (0.00157) (0.00161)

Real GDP per capita 0.0352* 0.148*** 0.0445* 0.0834* 0.183*** 0.577*** 0.0438** 0.172*** − 0.0210
(0.0180) (0.0315) (0.0244) (0.0488) (0.0648) (0.168) (0.0207) (0.0372) (0.0208)
Political institutions 0.0120* 0.0775*** 0.195*** 0.314*** 0.304*** − 0.0125 0.0897* 0.435*** 0.00260
(0.00723) (0.0260) (0.0388) (0.0485) (0.0292) (0.0154) (0.0476) (0.0707) (0.00804)
Constant 0.115 − 0.818*** 0.834*** − 0.176 − 0.904* 0.0817 0.151 − 0.809*** 0.770***
(0.149) (0.259) (0.202) (0.406) (0.530) (0.393) (0.168) (0.288) (0.230)

Time Fixed Effect Yes Yes Yes Yes Yes Yes Yes Yes Yes
10

Observations 686 525 161 686 525 161 686 525 161
Number of Group 46 45 23 46 45 23 46 45 23
Instrument 27 26 17 27 26 17 27 26 17
AR1 − 3.091 − 3.338 − 3.035 − 3.186 − 3.766 − 3.905 − 3.989 − 3.453 − 3.005
z (p-value) 0.001 0.001 0.000 0.001 0.001 0.000 0.001 0.001 0.000
AR2 − 0.627 0.680 0.691 − 0.154 − 0.600 − 0.318 − 0.215 0.65 0.310
z (p-value) 0.531 0.509 0.489 0.877 0.549 0.365 0.830 0.517 0.315
Sargan’s Test 27.94 22.32 15.27 15.37 18.42 16.79 22.77 25.69 22.21
p-value 0.359 0.387 0.301 0.816 0.433 0.321 0.291 0.610 0.206
Hansen’s Test 8.188 6.440 6.918 5.214 6.980 9.555 8.777 10.56 6.840

The Journal of Economic Asymmetries 29 (2024) e00355

Chi2 (p) value) 0.515 0.695 0.837 0.815 0.444 0.0520 0.458 0.307 0.824
Fisher-test 1154 1107.8 1707.1 2717 2373.60 2160.99 3125 3301.4 152128
P-value 0 0 0 0 0 0 0 0 0

Table 2 shows the effect of central bank regulatory independence on bank lending. All variables are described in Appendix A.
Robust standard errors in parentheses.
***p < 0.01, **p < 0.05, *p < 0.1.
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

Table 3
Interactive effects of electoral cycles and central bank regulatory independence on bank lending.
Interactions with— Monetary policy Independence Macro-prudential Independence Central Bank Independence

VARIABLES Model 18 Model 19 Model 20

Bank lending t-1 − 1.003*** − 1.003*** − 1.002***

(0.00682) (0.00695) (0.00765)
Electoral cycle t-1 0.0203* 0.0201* 0.0203*
(0.0111) (0.0105) (0.0111)
Electoral cycle 0.0627*** 0.169*** 0.153**
(0.0169) (0.0157) (0.0718)
Electoral cycle t+1 ¡0.103*** ¡0.144*** ¡0.780***
(0.0395) (0.0286) (0.118)
Electoral cycle t+2 ¡0.0253*** ¡0.0241** ¡0.0858***
(0.00906) (0.0121) (0.0202)
Monetary policy independence − 10.16**
(4.808)
Electoral cyclet-1 × Monetary policy independence 0.00197***
(0.000182)
Electoral cycle × Monetary policy independence ¡0.939**
(0.380)
Electoral cyclet+1 × Monetary policy independence − 0.0233
(0.0160)
Electoral cyclet+2 × Monetary policy independence − 0.601***
(0.172)
Macro-prudential independence − 12.11***
(4.298)
Electoral cyclet-1 × Macro-prudential independence − 0.0854***
(0.0272)
Electoral cycle × Macro-prudential independence ¡0.101***
(0.0217)
Electoral cycletþ1 × Macro-prudential independence 0.141**
(0.0559)
Electoral cycletþ2 × Macro-prudential independence − 1.787***
(0.671)
Central bank independence − 1.314***
(0.478)
Electoral cycle t-1 × Central bank independence 0.00134***
(0.000343)
Electoral cycle × Central bank independence ¡0.0283**
(0.0114)
Electoral cycle tþ1 × Central bank independence − 0.0136*
(0.00760)
Electoral cycle tþ2 × Central bank independence − 0.235**
(0.112)
Banking crisis − 104.1*** − 107.5*** − 120.9***
(5.639) (6.446) (7.223)
Bank concentration − 0.108*** − 0.117*** − 0.0977***
(0.00856) (0.00830) (0.00946)
Credit risk − 1.200*** − 0.551*** − 1.005***
(0.0554) (0.0787) (0.0704)
Foreign entry 21.95*** 24.29*** 26.21***
(3.369) (4.691) (3.304)
Inflation − 0.00302*** − 0.00124 − 0.00153**
(0.000824) (0.000824) (0.000693)
Real GDP per capita 1.164** 2.073*** 1.846***
(0.466) (0.489) (0.576)
Political institutions 23.74*** 23.37*** 14.34***
(1.752) (2.654) (2.095)
Time Fixed Effect Yes Yes Yes
Constant − 29.36*** − 12.26** − 7.035
(4.198) (5.400) (12.43)

Observations 736 690 734

Number of Group 48 48 48
Instrument 23 23 23
AR (1) z (p value) − 3.38 (0.001) − 3.50 (0.001) − 3.27 (0.001)
AR (2) z (p value) 0.63 (0.499) 0.65 (0.517) 0.61 (0.491)
Sargan Test OIR 2.502 (0.286) 0.266 (0.606) 0.195 (0.207)
Hansen Test OIR: 5.15 (0.272) 5.07 (0.397) 5.20 (0.270)
Chi2 (p) value)
Fisher 2808.65*** 2156.63*** 2160.99***
Net Effect (Electoral cycle) ¡0.3674*** 0.1513*** 0.1376***

11
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

Table 2 shows the interactive effect of electoral cycle and central bank regulatory independence on bank lending. All variables are described in Appendix A.
Robust standard errors in parentheses.
***p < 0.01, **p < 0.05, *p < 0.1.

over-identifying instruments are valid” is accepted, (Roodman, 2009). We apply Windmeijer (2005) correction to produce robust
standard errors because the two-step estimator has been shown to be biased without this correction. The error term of the model was
tested for its assumptions of normality, autocorrelation and homoscedasticity. GMM can be used without having diagnostic tests
because by its very nature it is designed to solve the problems of endogeneity, autocorrelation, and heteroscedasticity. In addition, the
error term’s test for autocorrelation and serial correlation is shown to test the null hypothesis that the error term’s first and second
orders are serially correlated. This means that failure to reject the null hypothesis that there is no second-order serial correlation
implies that the original error term is serially uncorrelated and the moment conditions are correctly specified (that is, the value of AR
(2) >0.05).

Table 4
Interactive effects of electoral cycles and central bank regulatory independence on bank lending in countries with weak and strong political
institutions.
Countries with weak political institutions Countries with strong political institutions

VARIABLES Model 21 Model 22 Model 23 Model 24 Model 25 Model 26

Bank lending t-1 0.856** 0.836** 0.836** 0.845** 0.850** 0.869**

(0.364) (0.369) (0.369) (0.367) (0.368) (0.364)
Electoral cycle 1.935** 1.675* 1.685* 0.227** 0.230*** 0.230***
(0.873) (0.911) (0.916) (0.0895) (0.0667) (0.0667)
Monetary policy independence − 0.00779*** 0.00869***
(0.000222) (0.000533)
Electoral cycle × Monetary policy 3.651* ¡0.3714***
independence
(2.115) (0.1390)
Macro-prudential independence 0.471*** 12.19**
(0.177) (5.767)
Electoral cycle × Macro-prudential 0.0645*** ¡5.566**
independence
(0.00243) (2.489)
Central bank Independence − 12.68*** 0.251*
(4.709) (0.135)
Electoral cycle × Central bank Independence 1.554* ¡0.334*
(0.820) (0.173)
Banking crisis − 7.403 9.082 10.96 − 4.498 − 10.32 − 11.92
(12.58) (15.10) (15.33) (13.74) (12.64) (12.91)
Bank concentration − 0.0217* − 0.0587** − 0.0587** − 0.0245** − 0.0402** − 0.0248*
(0.0116) (0.0240) (0.0240) (0.0122) (0.0180) (0.0128)
Credit risk − 0.119* − 0.144** − 0.144** − 0.111* − 0.0384 − 0.734
(0.0621) (0.0705) (0.0705) (0.0638) (0.0417) (1.499)
Foreign entry 1.084*** 0.540** 0.540** 1.065*** 2.449*** 1.303***
(0.400) (0.238) (0.238) (0.392) (0.884) (0.475)
Inflation − 3.443** − 4.597** − 4.597** − 3.621** − 5.367** − 3.906**
(1.599) (1.972) (1.972) (1.648) (2.290) (1.755)
Real GDP per capita 12.19* 12.00* 15.54** 18.27** 18.09** 18.22**
(6.466) (6.997) (7.671) (7.917) (8.411) (8.644)
Time Fixed Effect Yes Yes Yes Yes Yes Yes
Constant 163.7** 169.8*** 141.4** 311.2*** 234.3*** 174.5**
(78.64) (51.81) (54.85) (89.23) (70.10) (83.06)

Net effect 3.5892*** 1.6848*** 2.5312*** 0.05508*** ¡0.8654*** 0.04777***

Observations 313 346 344 325 358 358
Number of Group 35 35 35 31 31 31
Instrument 19 19 19 19 19 19
AR (1) z (p value) − 3.677 − 3.112 − 3.297 − 3.221 − 3.682 − 3.113
(0.007) (0.007) (0.007) (0.007) (0.007) (0.007)
AR (2) z (p value) − 1.56 (0.120) − 1.53 (0.126) − 1.56 (0.120) − 1.54 (0.124) − 1.53 (0.126) − 1.54 (0.123)
Sargan Test 15.71 37.65 22.81 37.65 18.54 22.30
(p-value) (0.152) (0.895) (0.297) (0.895) (0.206) (0.542)
Hansen Test OIR: 30.28 (0.603) 28.11 (0.709) 29.31 (0.651) 30.86 (0.574) 30.02 (0.616) 28.46 (0.693)
Chi2 (p) value)
Fisher 1098.19*** 1135.69*** 1519.73*** 796.26*** 2396.68*** 1401.12***

All variables are described in Appendix A.

12
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

4. Empirical results and discussions

4.1. Presentation of results

The study presents and descriptive statistics and pairwise correlation coefficient matrix in the Appendix. In all the panel estimates
presented in Tables 1–4, we fail to reject the null hypothesis of no autocorrelation for AR (2). Also, overidentifying restrictions were
valid for all instruments. We begin by presenting the results on the effect of electoral cycles on bank lending behaviour.

4.1.1. Impact of election on bank lending behaviour

We have hypothesized that electoral cycles are important in determining the levels of bank lending behaviour. To test this, we
introduce the electoral cycle variables into the regression to examine the impacts on bank lending. In Table 1, we present the estimates
of the effect of electoral cycles on bank lending behaviour (see models 1–8). First, we run the regression with the contemporaneous
election values but without the controls, and then, include the controls to obtain a clearer understanding. The model allows us to
introduce the past lending behaviour of banks and indicators of political institutions into the model. In Table 1, we find that past
lending behaviour of banks leads to lower lending behaviour in the subsequent year (models 1–8), and this is consistent throughout the
results. This does not agree with the work of Ladime et al. (2013), who found a positive relationship between past years’ lending
behaviour and current lending behaviour of banks. Our negative relationship between previous years’ lending and current lending
behaviour, is possible because of the aggressive lending behaviour of banks that may persist in the credit market and induce a future
reduction in the degree of lending. This agrees with Papademos (2009), who shows that the risk built up by banks in good times may
result in future restrictions on the supply of loan through its impact on capital.
As explained earlier, political business cycle describes how the government influences the economy in order to secure re-election. In
view of that, we introduce the contemporaneous, lag and lead values of election periods into the model and observe their impacts on
bank lending beheviour. In Table 1, we find that the contemporaneous effect of electoral cycle on bank lending beheviour is positive
and significant (models 2–8). This suggests that banks lend more during election periods. In support of the “opportunistic” model
(Iddrisu & Ebo Turkson, 2020) and the work of Koetter and Popov (2021), both the government in power and the left-wing government
would be more likely to increase social spending to increase their chances of winning the next-election, and hence be in higher need of
bank funding. Similarly, past-election dummy has a positive impact on bank lending behaviour. We observe that the first lag of
elections has a positive impact on bank lending – indicating that immediate past election periods (pre-election) increase bank lending
(see Table 1). It is obvious that the positive impacts of electoral cycles on bank lending are greater during election periods compared to
pre-election periods. In Table 1, we observe that the lead of election dummy in year 1 has a negative and significant effect on bank
lending. The negative impact is magnified two years after election. This suggests that banks reduce lending immediately after election
and the negative impact is persistent over the post-election era. This effect is intuitive and economically meaningful in the sense that
banks reduce aggressive lending behaviour after elections and the reason could be attributed to a reduction of a build-up risks and
perhaps the quest to build a resilient banking system from a sudden change in government. It also supports the work of Englmaier and
Stowasser (2017) who show that total lending by local savings banks is substantially higher during an election year, and declines
afterwards. Our results imply that the negative impact ofelection on bank lending behaviour persists before election years but dis­
sipates after elections.
In general, our results confirm that electoral cycles are important in determining the levels of bank lending behaviour in different
political business regimes. Specifically, it supports the hypothesis of Iddrisu and Ebo Turkson (2020) who provide evidence that
political business cycles increase pricing behaviour of banks in Africa. However, it disagrees with the findings of Leon and Weill (2022)
who show that firms are more credit constrained in election years and pre-election years as election exacerbate political uncertainty.
Thus, our results are in line with the ‘opportunistic behaviour theory’ which argues that all governments and institutions, irrespective
of their ideological orientation or political affiliation, apply expansionary policies ahead of elections in order to increase their
recognition and brighten their re-election chances. Therefore, commercial banks lend more around periods of electoral cycles but tend
to reduce lending in the long term over the political business cycle.

4.1.2. Impact of regulatory independence of central bank on bank lending behaviour

In this section, we test whether regulatory policy independence of the central bank has a significant effect on bank lending be­
haviours across electoral regimes. Table 2 shows the results of the individual regulatory independence of central bank (monetary,
macro-prudential and CBI) on bank lending, and the impacts across different electoral regimes. For instance, in Table 2, we show that
the response of bank lending to the individual regulatory independence of the central bank do not take effect instantaneously, and
therefore, for robustness purposes, we introduce both the lag and the lead values of the regulatory independence variables to capture
the instantaneous or contemporaneous effects. In addition, it is possible that the bank lending impacts of the dynamics of the individual
regulatory independence of the central bank, based on their lag and lead values – may vary across electoral events. For this reason, we
split the sample into periods of election and periods without election. We do this to examine the extent to which the effect of the
regulatory independence variables on bank lending differ across different electoral regimes.
In Table 2, for instance, we show the results for the full sample, as well as the split samples based on the respective periods of
election and periods without election. In Table 2, we observe that the contemporaneous effect of monetary policy independence on
bank lending was negative and significant (see, Model 9). This suggests that countries that enforce monetary policy autonomy of the
central bank induce a negative impact on bank lending behaviour. This has its roots from the lens of monetary policy transmission
mechanism, as argued by Friedman (1968), who explained that an increase in money supply leads to a decrease in the monetary policy

13
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

rate, with a resultant increase in bank lending. Therefore, contractionary monetary policy independence, for instance, an increase in
the monetary policy rates by the central banks, restricts the liquidity and the ability of banks to lend, thereby reducing credit expansion
to borrowers and business firms, as supported by Ciccarelli et al., 2015). This is in line with the findings of Modugu et al. (2022), Abuka
et al. (2015, 2019) and Borio and Gambacorta (2017) who supported the claim that an increase in monetary policy rates, decreases
bank lending behaviours in developing countries. After introducing the lag and forecast/forward values of monetary policy inde­
pendence, we observe that the negative impact is reduced for pre-implementation of monetary policy independence while it is
enhanced for the post implementation of monetary policy independence (see model 9). This is because the pre-implementation of
policy reforms relaxes the level at which bank lending is reduced. It is clear from our results that electoral cycles influence the extent of
impact of monetary policy dynamics on bank lending. For instance, the negative impacts of monetary policy independence, its lag and
lead values on bank lending are lower in periods without election (model 10) compared to periods with election (model 11).
In terms of macro-prudential independence, Table 2 shows that macro-prudential independence action of the central bank has a
positive and significant effect on bank lending behaviour (see, model 12). Macro-prudential policy actions by the central bank offer
banks the incentive to maintain capital in their buffer and reserves, and thus, banks have to increase their level of capital in response to
any increase in risk. According to the regulatory hypothesis, a positive relationship exists between the capital level and the risk-taking
incentives. Tightening of the macro-prudential policy independently by the central bank raises the stock of capital in banks’ reserves
and consequently, increases the ability of banks to create loans. Therefore, our findings suggest that stringent macro-prudential au­
tonomy of the central bank increases bank lending behaviour. This agrees with a recent study by Hodula and Ngo (2021) who provided
a robust estimates that a macro-prudential policy tightening leads to an increase in shadow bank lending. While the use of
macro-prudential measures could lead to higher capitalization and make the financial sector more resilient and reduce its risk
exposure. This contradicts the findings of Cehajic and Kosak (2022), who show that the implication could mean restricted lending to
firms, especially smaller firms with financing options and considerable reliance on bank credit. We show that the contemporaneous
and lag values of macro-prudential independence positively affect bank lending but lead values of macro-prudential independence
increases bank lending (see Model 12). This suggests that bank lending reduces substantially when countries implement
macro-prudential independent action in the future. Table 2 shows that macro-prudential independence increases bank lending in
periods without election while it reduces bank lending in election periods (Models 13 and 14). The positive impact of the lag and lead
values of macro-prudential independence on bank lending in periods without election (model 13) is relatively greater compared to
periods of election (model 14). The implication is that banks are more skeptical to take risk during election periods, despite the degree
of compliance to macro-prudential standards.
In Table 2 (Model 15), central bank independence and its lag have a positive and significant effect on bank lending behaviour. Our
findings support the claim by Agoba et al. (2020) that central bank independence promotes access to credit by the private sector by
reducing inflation (price stability). However, the positive impacts are reversed when countries implement central bank independence
in the future or in the subsequent year. This implies that countries that allow the central bank to set independent policy instruments in
the long term allow them to monitor the opportunistic behaviour of managers in the credit market, generate optimal returns that gives
them greater power to reduce lending. In addition, the positive impact of the central bank independence, its lag and lead values on
bank lending in periods without election (model 16) is reduced in periods of election (model 17). The implication is that the dynamics
of central bank independence policies induce an increase in bank lending behaviours across different electoral regimes, hence, require
effective regulatory measures by reserve banks.

4.1.3. Interactive effects of electoral cycle and regulatory independence of central bank on bank lending
It is also possible that central bank wants to signal that banks’ lending patterns during elections may not yield desirable outcome. In
view of that, we test whether regulatory policy independence is important in shaping the effect of electoral cycles on bank lending
behaviours. Thus, we include the interaction of electoral cycles with individual regulatory independence of the central bank. We
discuss the net effects from the interactions between the electoral cycles and the individual regulatory independence of the central
bank. Consistent with Brambor et al. (2006) on pitfalls of interactive regressions, we cannot establish policy implications exclusively
on marginal effects. Hence, we interpret the net effects (overall impact), as supported by Ghosh (2022) and Asongu and Nwachukwu
(2016). In Table 3, the unconditional effects of electoral cycles show that bank lending is increased before election and the extent of the
increment is greater during election years. However, bank lending declines after an election and the reductive effect falls in the
subsequent years after election.
The independent regulatory policies of the central bank are important toolkits for shaping the behaviours of bank lending in
electoral cycles. For instance, in Table 3, the unconditional effect of elections on bank lending is 0.0627 while the conditional effect
(coefficient of interaction term between electoral cycle and monetary policy independence) is − 0.939 (see model 18). This means that
the marginal effect of electoral cycle is negative when interacted with monetary policy independence. For better interpretation that is
consistent with Brambor et al. (2006), we compute the net effect of electoral cycle to be equal to − 0.3674 [0.0627 + (− 0.939 x mean of
monetary policy independence)] when the mean of monetary policy independence is 0.458. Thus, the negative net effect suggests that
monetary policy independence alters the positive impact of election on bank lending into a negative. Similarly, based on the marginal
effect (coefficient of the interaction term between the lag values of electoral cycle and monetary policy independence), the positive
impact of pre-election (first lag of electoral cycle) on bank lending is reduced when interacted with monetary policy independence
while the negative impact of the lead values of electoral cycle on bank lending is enhanced when interacted with monetary policy
independence.
In Table 3, the unconditional effect of elections on bank lending is 0.169 while the conditional effect (coefficient of interaction term
between electoral cycle and macro-prudential independence) is − 0.101 (see Model 19). This means that the marginal effect of electoral

14
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

cycle is negative when interacted with macro-prudential policy independence. For better interpretation, the computation of the net
effect (0.1513) of electoral cycle is less positive compared to the unconditional effect. This suggests that macro-prudential indepen­
dence reduces the positive impact of election on bank lending. Similarly, based on the marginal effect (i.e., the coefficient of the
interaction term between the lag values of electoral cycle and macro-prudential independence), the positive impact of pre-election
(first lag of electoral cycle) on bank lending is altered when interacted with macro-prudential independence. In addition, the nega­
tive impact of the lead values of electoral cycle (in year 1) on bank lending is reversed to positive when interacted with macro-
prudential independence, but the negative impact of the lead values of electoral cycle (in year 2) on bank lending is enhanced
when interacted with macro-prudential independence.
In Table 3, the unconditional effect of elections on bank lending is 0.153 while the conditional effect (coefficient of interaction term
between electoral cycle and central bank independence) is − 0.0283 (see model 20). This means that the marginal effect of electoral
cycle is negative when interacted with central bank independence. For better interpretation, the computation of the net effect (0.1376)
of electoral cycle is less positive compared to the unconditional effect. This suggests that central bank independence reduces the
positive impact of election on bank lending. Similarly, based on the marginal effect (i.e., the coefficient of the interaction term between
the lag values of electoral cycle and central bank independence), the positive impact of pre-election (first lag of electoral cycle) on bank
lending is reduced when interacted with central bank independence while the negative impacts of the lead 1 and lead 2 values of
electoral cycle on bank lending are reduced and enhanced respectively when both are interacted with central bank independence.
It can be deduced from the interactions that countries with strong monetary and macro-prudential autonomy are more likely to
tighten regulations within electoral cycles and during economic booms when credit growth are high (Funke et al., 2016; Antoniades &
Calomiris, 2018; Doerr et al., Voth, 2018; Gyongyosi & Verner, 2019). Regulatory policy is formulated by governments to impose
restrictions and controls on certain activities or behaviour while regulatory independence is a law that establishes a regulator as
independent authority to control certain behaviours of the banking sector (Lanneau, 2021). Although elections send good signal to
commercial banks to lend more to the government in power and other political parties in order to increase their chances of winning an
election, for countries with strong independent central bank regulatory policy, the central bank or independent regulatory authority
may impose restrictions on banks’ lending behaviours to reduce potential financial losses during election periods. This explains why
our overall impacts, based on the net effects, shows that independent central bank regulatory policies tame the nexus between electoral
cycle and bank lending.
In general, our results confirm that independent central bank regulatory policies (monetary, macro-prudential and central bank
independence) dampen the positive effects of elections on bank lending before and during elections while they amplify the reductive
effects of electoral cycle on bank lending immediately and years after the election.

4.2. Robustness results: Interactive effects across strong and weak political institutions

Our earlier findings indicate that banks reduce their lending in electoral cycles when interacted with the individual regulatory
independence of the central bank. It does not show the overall impacts of this behaviour across political institutions. The study shows
evidence of the net effect of elections on bank lending at levels of the regulatory independence indexes across the political institutions.
Following Brambor et al. (2006) and Asongu and Nwachukwu (2016), we compute the net effects. In Table 4, for instance, in models 21
and 24 respectively, it can be deduced that for countries with strong political institutions, the positive impact of elections on bank
lending is reduced at increasing levels (tightening) of independent monetary policy of the central bank while for countries with weak
political institutions, the positive impact of elections on bank lending is magnified at increasing levels (tightening) of independent
monetary policy of the central bank.
Similarly, in Table 4, for countries with a weak political institution, the net effect of electoral cycle is more positive at increasing
levels (tightening) of macro-prudential independence (see model 22) but for countries with strong political institutions, the net effect is
less positive at increasing levels (tightening) of macro-prudential independence (see models 25). In Table 4 (model 23), the net effect of
electoral cycle in countries with weak political institutions is more positive at increasing levels of central bank independence than the
net effect for countries with strong political institutions.
Although, elections generally send good signals to governments, political institutions, the financial sector and regulators about the
local environment (Agbloyor, 2019), for countries with poor political institutions, political interference may hinder or restrict the
independent function of central banks – which may induce greater lending behaviours of banks. Thus, central bank regulatory policy
independence magnify the extent to which banks on-lend during electoral events. This confirms our results that, banks in countries
with strong political institutions are able to reduce their lending activities at greater levels of regulatory independence index of the
central bank compared to countries in weak political institutions. As a policy implication, elections can consolidate the gains of bank
lending from independent central bank regulatory policies. However, the corresponding low net effects imply that democratic system
and electoral standards need to be improved in order to accelerate the underlying gains in the development of independent central
bank regulatory settings.

5. Conclusion and policy implications

The aim of the study is to examine the effect of regulatory independence of the central bank in shaping the impact of electoral cycles
on bank lending behaviour in Africa. By employing the dynamic system Generalized Method of Moments (SGMM) Two-Step estimator
for a panel dataset of 54 African countries over the period, 2004–2022, the study found that banks lend more when elections are getting
closer but the total lending by banks is substantially higher during election years, and declines afterwards. The study shows that

15
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

countries that enforce monetary policy autonomy of the central bank induce a negative impact on bank lending behaviour while those
that apply strong macro-prudential independent action and central bank independence reduce lending substantially in the long term.
The study provides evidence to empirically support that monetary, macro-prudential and central bank independence of the central
dampen the positive effect of elections on bank lending around election years while they amplify the reductive effects on bank lending
immediately and in periods after the election. In addition, banks in countries with strong political institutions are able to reduce their
lending activities in stringent regulatory independence of the central bank compared to countries in weak political institutions.
Further, although we do not test this directly, the results suggest that incumbents choose good policies (monetary policy, etc.),
contrary to the independent functions of central bank, in order to enhance their chances of success. Consequently, the environment for
bank lending activities is likely to be friendlier during elections, leading to greater lending by banks. The study has policy implications
in the sense that government, policymakers and political institutions should come up with appropriate policies that controls political
influence, and the actions of banks to lend excessively during elections. Thus, efforts of making elections an instrument for deepening
democratic process of a country will help reduce the country’s risk profile, and hence control the level of bank lending around electoral
cycles. Based on that, future studies should consider the threshold point at which banks can lend during election years.
In addition, policymakers should put forward the right policy mix between the individual regulatory policies of independent central
bank to control bank lending behaviours around election periods. Thus, these policies should be well defined to reflect a tightening or
loosening targeted policy instrument around electoral cycles, and during good times and bad times. Further, lending policies have
more immediate response and thus are politically sensitive from independent central bank’s regulatory policies in a country with
strong political institution compared to those in a weak political institution. Hence, there is a wake-up call for countries with weak
independent central bank policy and political institutions to strengthen their independent regulatory policy policies. This will enable
them better strategize to yield a desirable outcome of bank lending to the real economy during election years.
The findings in this study obviously leave space for future research, especially as it pertains to assessing if the established findings
withstand empirical scrutiny in other developing regions of the world. Consequently, it would should be interesting for future study to
consider other variables that capture the institutional architecture of the micro- and macro-prudential policies, the degree of inde­
pendence of prudential authorities, and whether the central bank is involved or not in the prudential supervision of the banking sector
and how these measures affect bank lending. Moreover, owing to data availability constraints at the time of the study, digital currency
variables are not involved and hence, considering these variables in future studies is worthwhile.

Ethical approval

All authors have read and approved the final manuscript.

Informed consent

Corresponding author is prepared to provide documentation of compliance with ethical standards and sent to the journal upon
request.

Availability of data and material

The datasets used and/or analysed during the current study are available (with corresponding author) on reasonable request.

Funding

This study received no financial support from any institution or person. The study was carried out as a result of collaborative efforts
by authors without any direct or indirect support from any institution.

CRediT authorship contribution statement

Daniel Ofori-Sasu: Conceptualization, Investigation, Supervision, Writing – original draft. Elikplimi Komla Agbloyor: Writing –
review & editing. Dennis Nsafoah: Investigation. Simplice A. Asongu: Conceptualization, Formal analysis, Investigation, Supervi­
sion, Validation, Writing – review & editing.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to
influence the work reported in this paper.

Data availability

Data will be made available on request.

16
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

Acknowledgements

The authors are grateful to the University of Ghana Business School for funding this project. We are also thankful to the participants
of the University of Ghana Business School PhD seminar series for their criticism and comments which helped improve the quality of
this paper. We are grateful to the Carnegie Corporation of New York (CCNY) BANGA Project to the University of Ghana and the
University of Ghana Business School for funding this project. All errors and omissions remain ours.

Appendix A. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.jeca.2024.e00355.

References

Abor, J. Y., Agoba, A. M., Mumuni, Z., & Yawson, A. (2022). Monetary policy, central banks’ independence, and financial development in Africa. In The economics of
banking and finance in Africa (pp. 227–267). Cham: Palgrave Macmillan.
Abuka, C., Alinda, R. K., Minoiu, C., Peydro, J.-L., & Presbitero& Presbitero, A. F. (2019). Monetary policy and bank lending in developing countries: Loan
applications, rates, and real effects. Journal of Development Economics, 139, 185–202.
Agarwal, S., Morais, B., Ruiz Ortega, C., & Zhang, C. (2016). The political economy of bank lending: Evidence from an emerging market. World Bank Policy Research
Working Paper, 7577.
Agbloyor, E. K. (2019). Foreign direct investment, political business cycles and welfare in Africa. Journal of International Development, 31(5), 345–373.
Agoba, A. M., Abor, J., Osei, K. A., & Sa-Aadu, J. (2017). Central bank independence and inflation in Africa: The role of financial systems and institutional quality.
Central Bank Review, 17(4), 131–146.
Agoba, A. M., Abor, J. Y., Osei, K. A., & Sa-Aadu, J. (2020). Do independent central banks exhibit varied behaviour in election and non-election years?: The case of
fiscal policy in Africa. Journal of African Business, 21(1), 105–125.
Akinci, O., & Olmstead-Rumsey, J. (2018). How effective are macroprudential policies? An empirical investigation. Journal of Financial Intermediation, 33, 33–57.
Ali, F., Saleem, A., Zahid, R. M. A., Barczi, J., & Sagi, J. (2022). Lending behaviours of banks during elections and business cycles: One-step system gmm approach.
Academy of Accounting and Financial Studies Journal, 26(5), 1–16.
Altunbas, Y., Binici, M., & Gambacorta, L. (2018). Macroprudential policy and bank risk. Journal of International Money and Finance, 81, 203–220.
Amidu, M. (2006). The link between monetary policy and banks lending behaviour: The Ghanaian case. Banks and Bank Systems, 1(N◦ 4), 38–48.
Arellano, M., & Bover, O. (1995). Another look at the instrumental variable estimation of error-components models. Journal of Econometrics, 68(1), 29–51.
Arnone, M., Laurens, B. J., Segalotto, J. F., & Sommer, M. (2009). Central bank autonomy: Lessons from global trends. IMF Staff Papers, 56(2), 263–296.
Asongu, S. A., & Nwachukwu, J. C. (2016). The role of governance in mobile phones for inclusive human development in Sub-Saharan Africa. Technovation, 55, 1–13.
Asongu, S. A., & Odhiambo, N. M. (2021). Remittances and value added across economic sub-sectors in Sub-Saharan Africa. Quality and Quantity, 1–19.
Auer, R., Matyunina, A., & Ongena, S. (2022). The countercyclical capital buffer and the composition of bank lending. Journal of Financial Intermediation, Article 100965.
Ayyagari, M., Beck, T., & Peria, M. S. M. (2017). Credit growth and macroprudential policies: Preliminary evidence on the firm level. In B. for International
Settlements (Ed.), Financial systems and the real economy, volume 91 of BIS papers chapters (pp. 15–34). Bank for International Settlements.
Azzimonti, M. (2019). Partisan conflict, news, and investors’ expectations’ (Working Paper No. 2019-07). Stony Brook University.
Balls, E., Howat, J., & Stansbury, A. (2018). Central Bank independence revisited: After the financial crisis, what should a model central bank look like?M-RCBG associate
working paper series | No. 87.
Behncke, S. (2022). Effects of macroprudential policies on bank lending and credit risks. Journal of Financial Services Research, 1–25.
Bertay, A. C., Demirgüç-Kunt, A., & Huizinga, H. (2015). Bank ownership and credit over the business cycle: Is lending by state banks less procyclical? Journal of
Banking & Finance, 50, 326–339.
Bircan, Ç., & Saka, O. (2018). Political lending cycles and real outcomes: Evidence from Turkey. EBRD and LSE working paper. https://www.qeh.ox.ac.uk/sites/www.
odid.ox.ac.uk/files/Bircan_Saka-Political%20lending%20cycles%20in%20Turkey.pdf. (Accessed 1 July 2022).
Bircan, Ç., & Saka, O. (2019). Elections and economic cycles: What can we learn from the recent Turkish experience?. https://openaccess.city.ac.uk/id/eprint/26785/1/.
(Accessed 1 July 2022).
BIS. (2017). Macroprudential frameworks, implementation and relationship with other policies. Number 94 in BIS Papers. Bank for International Settlements https://
papers.ssrn.com/sol3/papers.cfm?abstract_id=3101175. (Accessed 1 July 2022).
Block, S. A. (2002). Political business cycles, democratization, and economic reform: The case of Africa. Journal of Development Economics, 67(1), 205–228.
Bodenstein, M., Guerrieri, L., & LaBriola, J. (2019). Macroeconomic policy games. Journal of Monetary Economics, 101, 64–81.
Borio, C., & Gambacorta, L. (2017). Monetary policy and bank lending in a low interest rate environment: Diminishing effectiveness? Journal of Macroeconomics, 54,
217–231.
Brambor, T., Clark, W. R., & Golder, M. (2006). Understanding interaction models: Improving empirical analyses. Political Analysis, 14(1), 63–82.
Brei, M., & Schclarek, A. (2013). Public bank lending in times of crisis. Journal of Financial Stability, 9(4), 820–830.
Buch, C., Vogel, E., & Weigert, B. (2018). Evaluating macroprudential policies. ESRB: Working paper series No. 2018/76. . (Accessed 1 July 2022).
Carrillo, J. A., Mendoza, E. G., Nuguer, V., & Roldán-Peña, J. (2021). Tight money-tight credit: Coordination failure in the conduct of monetary and financial policies.
American Economic Journal: Macroeconomics, 13(3), 37–73.
Carvalho, D. (2014). The real effects of government-owned banks: Evidence from an emerging market. The Journal of Finance, 69(2), 577–609.
Cehajic, A., & Kosak, M. (2021). Tightening and loosening of macroprudential policy, its effects on credit growth and implications for the COVID-19 crisis. Economic
and Business Review, 23, Article 207e233.
Cehajic, A., & Kosak, M. (2022). Bank lending and small and medium-sized enterprises’ access to finance – effects of macroprudential policies. Journal of International
Money and Finance, 1–23.
Chen, Y. S., Shen, C. H., & Lin, C. Y. (2014). The benefits of political connection: Evidence from individual bank-loan contracts. Journal of Financial Services Research,
45(3), 287–305.
Ciccarelli, M., Maddaloni, A., & Peydró, J. L. (2015). Trusting the bankers: A new look at the credit channel of monetary policy. Review of Economic Dynamics, 18(4),
979–1002.
Cohen, B. H., & Edwards, G. (2017). The new era of expected credit loss provisioning. BIS Quarterly Review. March https://papers.ssrn.com/sol3/papers.cfm?abstract_
id=2931474. (Accessed 1 July 2022).
Cukierman, A. (2008). Central bank independence and monetary policymaking institutions - past, present and future. European Journal of Political Economy, 24,
722–736.
Cukierman, A. (2013). Regulatory reforms and the independence of central banks and financial supervisors. In E. G. Morten Balling, States (Ed.), Banks and the
financing of the economy:monetary policy and regulatory regimes (pp. 121–134). Vienna: SUERF.
Czaplicki, M. (2022). Measuring the restrictiveness of (macro) prudential policy: The case of bank capital regulation in Poland. Journal of Banking Regulation, 23(3), 322–338.

17
D. Ofori-Sasu et al. The Journal of Economic Asymmetries 29 (2024) e00355

De Waal, A., Gupta, R., & Jooste, C. (2018). South Africa’s monetary policy independence: Evidence from a global new-keynesian DSGE model. Applied Economics
Letters, 25(12), 840–846.
Dinç, I. S. (2005). Politicians and banks: Political influences on government-owned banks in emerging markets. Journal of Financial Economics, 77(2), 453–479.
Doumpos, M., Gaganis, C., & Pasiouras, F. (2015). Central bank independence, financial supervision structure and bank soundness: An empirical analysis around the
crisis. Journal of Banking & Finance, 61, S69–S83.
Dreher, A., Lang, V. F., & Richert, K. (2019). The political economy of International Finance Corporation lending. Journal of Development Economics, 140, 242–254.
Duff, A. W. (2014). Central bank independence and macroprudential policy: A critical look at the US financial stability framework. Berkeley Bus. LJ, 11, 183.
Edge, R. M., & Liang, N. (2019). New financial stability governance structures and central banks. SSRN Working Paper.
Englmaier, F., & Stowasser, T. (2017). Electoral cycles in savings bank lending. Journal of the European Economic Association, 15(2), 296–354.
Epure, M., Mihai, I., Minoiu, C., & Peydró, J.-L. (2017). Household credit, global financial cycle, and macroprudential policies: Credit register evidence from an
emerging country. Working Papers, 1006.
Fang, H., Li, Z., Xu, N., & Yan, H. (2018). In the shadows of the government: Relationship building during political turnovers (No. w25300). National Bureau of Economic
Research.
Farhi, E., & Werning, I. (2014). Dilemma not trilemma? Capital controls and exchange rates with volatile capital flows. IMF Economic Review, 62(4), 569–605.
Friedman, H. (1968). Magnitude of experimental effect and a table for its rapid estimation. Psychological Bulletin, 70(4), 245.
Funashima, Y. (2016). The Fed-induced political business cycle: Empirical evidence from a time–frequency view. Economic Modelling, 54, 402–411.
Fungáčová, Z., Schoors, K. J., Solanko, L., & Weill, L. (2020). Political cycles and bank lending in Russia.
Gabriel, V. J., Lazopoulos, I., & Lima, D. (2022). Institutional arrangements and inflation bias: A dynamic heterogeneous panel approach. Journal of Money, Credit, and
Banking.
Gambacorta, L., & Murcia, A. (2017). The impact of macroprudential policies and their interaction with monetary policy: An empirical analysis using credit registry data.
Gerschenkron, A. (2015). Economic backwardness in historical perspective (1962). Cambridge MA.
Ghosh, S. (2022). Elections and provisioning behaviour: Assessing the Indian evidence. Economic Systems, 46(1), Article 100943.
Gyeke-Dako, A., Agbloyor, E. K., Agoba, A. M., Turkson, F., & Abbey, E. (2022). Central Bank independence, inflation, and poverty in Africa. Journal of Emerging
Market Finance, 21(2), 211–236.
Hodula, M., & Ngo, N. A. (2021). Does macroprudential policy leak? Evidence from non-bank credit intermediation in EU countries. Czech National Bank. Working Paper
Series ——— 5/2021.
Hussain, M., & Bashir, U. (2019). Impact of monetary policy on bank lending: Does market structure matter? International Economic Journal, 33(4), 620–648.
Iddrisu, A. G., & Ebo Turkson, F. (2020). Political business cycles, bank pricing behaviour and financial inclusion in Africa. Cogent Economics & Finance, 8(1), Article
1762286.
Jeanneney, S. G. (2006). Independence of the Central Bank of West African states: An expected reform? Revue d’Économie du Développement, 14(1), 45–77.
Jiménez, G., Ongena, S., Peydró, J.-L., & Saurina, J. (2017). Macroprudential policy, countercyclical bank capital buffers, and credit supply: Evidence from the
Spanish dynamic provisioning experiments. Journal of Political Economy, 125(6), 2126–2177.
Jones, E. (2022). Do central bankers dream of political union? From epistemic community to common identity. In Constructing the EU’s political identity (pp. 71–88).
Cham: Palgrave Macmillan.
Kakes, J., & Sturm, J. E. (2002). Monetary policy and bank lending:: Evidence from German banking groups. Journal of Banking & Finance, 26(11), 2077–2092.
Kanga, D. (2021). Banks and monetary policy transmission in the West African economic and monetary union-research paper. https://www.africaportal.org/publications/
banks-and-monetary-policy-transmission-west-african-economic-and-monetary-union-research-paper/. (Accessed 1 July 2022).
Keita, K., & Turcu, C. (2022). Promoting counter-cyclical fiscal policy: Fiscal rules versus institutions. Comparative Economic Studies, 1–46.
Klein, M. W., & Shambaugh, J. C. (2015). Rounding the corners of the policy trilemma: Sources of monetary policy autonomy. American Economic Journal:
Macroeconomics, 7(4), 33–66.
Klüh, U., & Urban, J. (2022). Independently green? An integrated strategy for a transformative ECB (No. 9). ZNWU Discussion Paper.
Koetter, M., & Popov, A. (2021). Political cycles in bank lending to the government. Review of Financial Studies, 34(6), 3138–3180.
Kohn, D. (2015). Implementing macroprudential and monetary policies: The case for two committees. Speech at the Federal Reserve Board’s Boston Conference.
Krishnamurti, D., & Lee, Y. C. (2014). Macroprudential policy framework: A practice guide. World Bank Publications.
Kumar, N. (2020). Political interference and crowding out in bank lending. Journal of Financial Intermediation, 43, Article 100815.
Martinez-Resano, J. R. (2004). Central bank financial independence.
Masciandaro, D., & Romelli, D. (2018). Central bankers as supervisors: Do crises matter? European Journal of Political Economy, 52, 120–140.
Meriläinen, J. M. (2016). Lending growth during the financial crisis and the sovereign debt crisis: The role of bank ownership type. Journal of International Financial
Markets, Institutions and Money, 41, 168–182.
Meuleman, E., & Vander Vennet, R. (2020). Macroprudential policy and bank systemic risk. Journal of Financial Stability, 47, Article 100724.
Micco, A., & Panizza, U. (2006). Bank ownership and lending behaviour. Economics Letters, 93(2), 248–254.
Mishra, P., Montiel, P., Pedroni, P., & Spilimbergo, A. (2014). Monetary policy and bank lending rates in low-income countries: Heterogeneous panel estimates.
Journal of Development Economics, 111, 117–131.
Modugu, K. P., & Juan Dempere, J. (2022). Monetary policies and bank lending in developing countries:evidence from Sub-Sahara Africa. Journal of Economics and
Development, 24(3), 217–229.
Mosley, P., & Chiripanhura, B. (2016). The African political business cycle: Varieties of experience. Journal of Development Studies, 52(7), 917–932.
Mpatswe, M. G. K., Tapsoba, M. S. J. A., & York, M. R. C. (2011). The cyclicality of fiscal policies in the CEMAC region. International Monetary Fund.
Müller, K. (2019). Electoral cycles in macro-prudential regulation. ESRB: Working Paper Series No. 2019/106.
Mwankemwa, L. P., & Mlamka, B. (2022). Effects of monetary policy on Bank’s credit dynamics in Tanzania. African Journal of Economic Review, 10(2), 13–39.
Nguyen, N. T., Wereley, S. T., & Shaegh, S. A. M. (2019). Fundamentals and applications of microfluidics. Artech house.
OECD. (2014). The Governance of regulators, OECD best practice principles for regulatory policy. Paris: OECD Publishing.
OECD. (2017). Creating a Culture of independence: Practical guidance against undue influence, the governance of regulators (p. 6). Paris: OECD Publishing.
Ozili, P. K. (2019). Bank loan loss provisioning during election years: Cross-country evidence. International Journal of Managerial Finance, 16(4), 413–431.
Paoli, B. D., & Paustian, M. (2017). Coordinating monetary and macroprudential policies. Journal of Money, Credit, and Banking, 49(2–3), 319–349.
Papademos, L. (2009). ECB financial stability review june 2009–opening remarks.
Rodriguez, J., & Santiso, J. (2008). Banking on democracy: The political economy of international private bank lending in emerging markets. International Political
Science Review, 29(2), 215–246.
Romelli, D. (2022). The political economy of reforms in central bank design: Evidence from a new dataset. Economic Policy, 37(112), 641–688.
Satragno, L. (2022). Monetary stability as a common concern in international law: Policy cooperation and coordination of central banks. Brill.
Sever, C., & Yücel, E. (2022). The effects of elections on macroprudential policy. Journal of Comparative Economics, 50(2), 507–533.
Shleifer, A., & Vishny, R. W. (1994). Politicians and firms. Quarterly Journal of Economics, 109(4), 995–1025.
Shrum, W., Chompalov, I., & Genuth, J. (2001). Trust, conflict and performance in scientific collaborations. Social Studies of Science, 31(5), 681–730.
Strong, C. O. (2021). Political influence, central bank independence and inflation in Africa: A comparative analysis. European Journal of Political Economy, 69, Article 102004.
Thomson, I. (2020). A literature review on regulatory independence in Canada’s energy systems: Origins, rationale and key features. University of Ottawa. https://ruor.
uottawa.ca/handle/10393/42482. (Accessed 1 July 2022).
Wintoki, M. B., Linck, J. S., & Netter, J. M. (2012). Endogeneity and the dynamics of internal corporate governance. Journal of Financial Economics, 105(3), 581–606.
Yun, Y., & Cho, B. (2022). Does monetary policy affect bank lending to households and firms differently? Economic Modelling, 109, Article 105783.

18