Social Sciences & Humanities Open 8 (2023) 100552

Contents lists available at ScienceDirect

Social Sciences & Humanities Open

journal homepage: www.sciencedirect.com/journal/social-sciences-and-humanities-open

Regular Article

Effect of firm size and corruption on financial challenges of savings and

loans companies: Evidence from Ghana
Eric Amankwaah a, *, Nana Okyir Baidoo b
a
Ghana Communication Technology University, PMB 100, Accra North, Ghana
b
University of Ghana, P.O Box LG 1181, Legon, Accra, Ghana

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

Keywords: Savings and loan companies have attempted to satisfy the needs of their customers by raising assets and giving
Firm size them credit. The study examines the financial challenges of savings and loan companies in Ghana from 2016 to
Corruption 2021. The data came from the Savings and Loans Companies database and the Bank of Ghana (BoG) database.
Financial challenges
Using panel data analysis and the Whited-Wu Index, the results show that the firm shows a consistent higher level
Savings and loans company
of financial constraints of − 1.7 in 2016. From 2019 to 2021, the company will have easier access to funds
Private credit
Collateral security compared to the period between 2017 and 2018. The study shows that cash holdings and private credit have a
Cashholdings positive influence on the financial constraints of the companies. The study also found that firm size, collateral
security, corruption index, and exchange rate significantly affect the financial challenges of companies nega­
tively. It turned out that there was no connection between Ghana’s economic growth and the company’s
financial challenges. The absence of sufficient financing significantly affects operations and profitability for the
company. It is therefore recommended that there be business realignment at the savings and loan companies in
Ghana. Company realignment is mostly done to get out of lines of business with poor margins and go into lines
that are more naturally financially smart and boost performance.

1. Introduction In the current environment, savings and loan businesses have made
an effort to meet the demands of their clients by building assets and
Ghana’s financial landscape has seen a significant transformation in extending loans. Nevertheless, they have not had the opportunity to
recent years with regard to the various institutions and zones of medi­ sufficiently increase their wealth by gathering enough institutional
ation. The situation has become uncertain, which has led to a decline in capital to finance irrecoverable fixed resources, providing a cushion to
the Ghanaian economy (Odoom et al., 2019). This raises challenging retain losses and reduce client deposits (Hassan, 2011; Olando et al.,
questions concerning the persistence, growth, and maintenance of 2015; Feleke, 2016).
practicable development (Musso & Schiavo, 2008; Abuhommous, Although it is acknowledged that S&LCs have considerably reduced
2016). Savings and loan companies (S&LCs) have been collapsing and poverty, there are still issues with its development that need to be
disappearing at an alarming rate recently, and the Bank of Ghana addressed. When the sector’s prospects are properly utilised, it may
liquidated several financial institutions around the country in 2008. successfully contribute to a feasible strategy to end poverty in Ghana
(Belnye, 2011). But the problem has persisted since 2013 and will (Bank of Ghana, 2007). The collapse of a number of MFIs in Ghana
continue to have disastrous effects. Thirty MFIs failed in Ghana during highlights important issues regarding the difficulties that institutions
the first quarter of 2013 due to an alleged inability to “sustain their must face in performing their tasks. Despite several studies, including
operations.” In a related incident, the Bank of Ghana (BoG) withdrew Obour (2013) and Odoom et al. (2019), the significant economic im­
the licences of 23 insolvent finance houses, businesses, and savings and pacts of S&LCs have once again been proven. They were unable to
loan organisations on Friday, August 16, 2019. In cases where a bank’s analyse the industry’s financial challenges. More specifically, the inap­
insolvency is confirmed, Bank of Ghana (BOG) has declined to approve plicability of studies conducted on S&LCs by Bates et al. (2009), Obour
the institution. These steps were taken early this year in compliance with (2013), and Odoom et al. (2019) is limited by their inability to clearly
section 123 (1) of the Banks and SDI Act, 2016 (Act 930) (Belnye, 2011). identify the financial challenges such institutions encounter in their

* Corresponding author.
E-mail address: eamankwaah@gctu.edu.gh (E. Amankwaah).

https://doi.org/10.1016/j.ssaho.2023.100552
Received 1 March 2023; Received in revised form 15 April 2023; Accepted 4 May 2023
Available online 9 June 2023
2590-2911/© 2023 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

operations. The glaring discrepancy between these findings may un­ development remained vital, given the cyclical nature of the economy
dermine efforts to explain operational problems and the effects of S&LCs and the emergence of a legal framework. Beck and Demirguc-Kunt
on the economy. (2006) noted that the degree of property rights strongly influences
Studies have shown that microfinance has greatly reduced poverty, both small and large enterprises’ needs for external finance. The finan­
so it is important to solve the development challenges it raises. A number cial struggles of organisations should be alleviated through improved
of studies, including Obour (2013) and Odoom et al. (2019), once more financial and legal growth. The financial heft of organisations will
demonstrated SLCs’ significant economic contributions to Ghana. These decline in a nation with stronger legal regulations. Financial specialists,
studies could not dissect the financial challenges within the sector. More for instance, will feel more certain that their financial contributions will
explicitly, the usefulness of studies led by Efthyyoulou & Vahter (2015), be secure. Speculators’ willingness to fund organisations is dependent
Obour (2013), and Odoom et al. (2019) on SLCs is constrained because on the level of legal security provided to them (Porta et al., 1997).
they cannot unmistakably define the financial challenges that in­ Businesses in nations with weak financial development invested multi­
stitutions face in their activities. The obvious gap between these in­ ple times as much in capital as those in nations with good financial
vestigations conceivably requires further study to explain operational development (Love, 2003). Demirguc-Kunt and Maksimovic assert that
issues and the ramifications of SLCs in the economy. Moreover, the the degree of financial middle-class growth significantly influenced how
inability to adequately highlight the financial challenges faced by SLCs organisations were financed externally.
complicates efforts to better place SLCs in the broader spectrum of na­ Financial requirements are becoming less important, especially for
tional development and financial intermediation roles. private investments, as a result of growing competition among banks
Most studies (Musso & Schiavo, 2008; Espinosa, 2015; Feng, 2011; and the establishment of locally held banks (Clarke et al., 2003). The
Argawi, 2014; Kifle, 2014; Ergetew, 2014) have focused on the con­ potential of banks to cause financial difficulties will increase as the
ceptual problems underlying the challenges facing manufacturing economy becomes increasingly dependent on them (as in Germany and
companies and have also focused on SMEs. Similarly, empirical studies Japan, for example). Ryan et al.’s (2014) findings support the market
to measure the financial challenges of savings and loan companies are control hypothesis. The relationship between the convergence of banks
very rare. This is due to the low presence of efficient savings and credit and the main financial imperatives has also been discussed by Beck,
companies on this continent and, as a result, the lack of data. Hence, the Demirguc-Kunt, and Maksimovic (2004), although only for all mone­
study provides empirical evidence on how firm size and corruption are tarily and institutionally developed nations. The availability of adjacent
influencing the financial challenges of savings and loan companies in external banks, a high level of institutional development, and a robust
Ghana. credit vault all undermine the link between bank focus and financial
The study is set up as follows: The study’s supporting literature is requirements (Beck et al., 2004).
provided in Section 2. The procedure and resources for the study are According to Jakob and Kostet (2016), financial needs are calculated
described in Section 3. Section 4 gives a data description, and empirical using a variable that considers the apparent level of obstacles businesses
findings are provided in Section 2. The procedure and resources for the face in their current operations, and the determining factor for financial
study are described in Section 3. Section 4 gives a data description and plausibility is whether or not businesses have undergone an examina­
presents empirical findings. The final section makes some concluding tion. The World Bank’s standard of legal conduct serves as our guiding
observations. principle. Regression analysis reveals a strong inverse link between
financial requirements and financial credibility. This illustrates that
2. Literature review increased financial disclosure veracity results in lower financial de­
mands on businesses. The results have no bearing on the standard of the
Many earlier investigations have discovered a connection between a law’s directing influence. In any event, we see that increased financial
nation’s degree of development and the limitations imposed on orga­ plausibility leads to tiny improvements in access to an external account
nisations (Desai et al., 2003; Love 2003; Beck & Demirguc-Kunt 2006; when the degree of rule of law is strong.
Beck 2007). The majority of research focuses on global constraints that Feng (2011) studied the effects of financial challenges on enterprises’
impede corporate development. Other studies just concentrate on fixed interests in China between 1998 and 2005. Using an Euler-
financial barriers. The extent of financial requirements for business and condition investment model, they find convincing evidence that a
national development appears to be a never-ending cycle: poor business “loan tendency” is at play. Exclusive corporations have greater financial
conditions make it harder for organisations to meet their financial ob­ obligations than state-owned and publicly traded companies. The evi­
ligations, and harder financial obligations have an influence on the dence also suggests that public companies are better off financially than
growth and development of the economy because businesses perform unlisted companies. The results also demonstrate that enterprises’ per­
worse. Gorodnichenko & Schnitzer (2011) found that national organi­ ceptions of financial constraints are lessened by the proximity of outside
sations’ capacity to grow and thereby penetrate more developed econ­ ownership. Second, we evaluate the effects of financial requirements on
omies was constrained by financial issues. According to their argument, businesses’ stock interests in China with a focus on firm heterogeneity
businesses in developing nations cannot invest in innovative initiatives using a mistake adjustment model extended with income. We find that
that may prevent them from getting better because of their financial the stock investment of exclusive businesses, businesses with outside
troubles. ownership, and businesses without political ties to the central or local
Corruption may really prevent a corporation that is having financial governments is significantly influenced by income. The results also
problems from expanding (Beck & Demirguc-Kunt, 2006; Chan, 2009; suggest that during the course of the study, enterprises’ awareness of
Love, 2003). According to Chan (2009), a 1 percent rise in insolvencies financial constraints grew in scope.
is correlated with a 0.22 to 0.26 percent drop in financially troubled According to Adden’s (2014) research, financially constrained or­
enterprises. Businesses experience fewer financial constraints in nations ganisations’ investments have greater benefit-related scrutiny than
with stronger legal frameworks. This includes laws, regulations, and flexible firms’ investments. This study also reveals that this result is
organisations that support the creation, registration, and implementa­ associated with differences in future write-downs and altruistic imped­
tion of guarantees, just like a strong corruption framework. The results ances, suggesting that the difference in tirelessness in productivity be­
of Love (2003) showed that the efficacy of the legal system, the reduc­ tween the two groups of enterprises is associated with differences in
tion of seizure risk, and the elimination of defilement had a substantial investment quality. Finally, the fact that investment in financially con­
bearing on the removal of financial restrictions on both large and small strained firms is associated with higher one-year odd returns than in­
enterprises. vestment in adaptable firms indicates that speculators do not fully
After limiting the company’s size, the outcomes for financial comprehend the role that financial requirements play in the link

2
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

between an endeavour and future profitability. were estimated using the theoretical framework presented in this study.
Espinosa’s (2015) found that Jordanian businesses must meet The following models were adopted to estimate the effects of firm size
financial obligations, but these obligations don’t appear to be related to and corruption on financial challenges:
any features of the businesses. The results also indicate that Jordanian
FC = β0 + β1 FS + β2 COR + β3 COL + β4 PC + β5 GDP + β6 EXR + ε (1)
businesses support their investments with obligations rather than
values. Access to outside financial resources, according to Musso and where:
Schiavo (2008), has a favourable effect on how transactions, capital FC denotes the financial Challenges.
stock, and work grow for businesses, and financial requirements are FS denotes Firm size,
unmistakably linked with short-term profitability growth. Financial COL denotes Corruption Index.
constraints completely increase the likelihood of leaving the market. COR - Collateral Security.
This final result is interpreted by us as a signal that obligated enterprises PC – Private credit from bank.
must reduce expenses in order to produce the assets they cannot raise in GDP – Gross domestic product.
financial markets. EXR-exchange rate
To assess the diversity of the effects of budgetary limitations, β is coefficient of the variables
Efthyyoulou and Vahter (2015) suggest that the impact of direct finan­ ε = Error term.
cial boundary proportions varies by operational and administrative The model assumes the error term is uncorrelated with the exoge­
components as well as by the financial orientation of the organization. In nous variables, meaning that the expected value of (xit εit) is equal to
particular, it appears that the creativity division has been more nega­ zero. Furthermore, it assumes the errors to be autonomously and nor­
tively impacted by financial needs than the administrative sector. Non- mally distributed, as well as independent of the lagged regressor, so the
exporters respond more rationally to financial obligations than other expected value of (Cit-1 εit) is zero. It is also believed that the explanatory
types of businesses. variables are correlated with the individual effect, leading to a non-zero
Wang (2016) examined how transactions were impacted by financial expected value of (xit vi). The firm-specific effect and the disturbance
challenges in 26 developing countries between 2001 and 2013. The error are considered to be independent, so E (vi it) and E (vi xit) are
study used a firm-level informational index to find that forced enter­ equal to zero. Finally, it is required to capture the dynamic impact of the
prises are 61.5 percent less likely to send out, compelled exporters trade dependent variable, as the past result will most likely influence the
74.2 percent less, and easing up financial restrictions increases firms’ present, leading to a non-zero expected value of Cit Ci,t-1.
fares. Aside from the existing fare pricing systems, developing countries
must find ways to improve their financial positions. The approach re­ 3.2.2. Description of variables and statement of hypotheses
stricts what corporations may gain due to a flawed agreement’s
enforceability. Companies use retained income to accumulate cash,
3.2.2.1. Financial challenges. To evaluate the degree of financial re­
relax financial limitations, and begin sending. To address the endoge­
strictions imposed on businesses, the Whited-Wu index was adopted. In
neity issue of financial imperatives, different ages and financial cir­
devising this index, Whited and Wu (2006) initiated from a model that
cumstances might be used.
proposes that external financing requirements affect the exchange of
immediate investment for delayed investment through the virtual
3. Material and methods
evaluation of restricted external resources. The index was formulated as
follows:
3.1. Research design
− 0.091CFit – 0.062DIVPOSit + 0.021TLTDit – 0.044LNTAit + 0.102ISGit
Quantitative research strategies focus on number management − 0.035SGit
techniques and everything that can be quantified effectively by exam­
(2)
ining phenomena and their relationships (Leedy 1993). To determine
the association between financial problems and company success, this where:
survey was done using ordinary least squares regression analysis. A more CFit is a measure of cashflow in comparison to all of a firm’s assets.
thorough examination of the relationship was conducted as part of an DIVPOSit indicates whether a company distributes money to share­
intriguing inquiry that would outline the historical background of the holders or not, with a value of 1 representing a dividend and 0 indica­
correlations between variables and make clear the goals of this tingno dividend.
relationship. TLTDitmeasures the portion of long-term debt relative to all the
firm’s assets.
3.2. Data collection and source LNTAit stands for the natural log of the firm’s total assets,
while ISGit and SGit refer to the three-digit industry and sales growth
Research information comes from secondary sources. With respect to of the business, respectively.
quantitative data, this study adopted secondary data because the in­ Compared to the Kaplan and Zingales (KZ) list of 1997, which was
formation can be collected and examined effectively. Panneerselvam frequently used to measure the degree of restrictions, Whited-Wu (2006)
(2006) stated that the collection of secondary data entails significant has two noteworthy benefits: avoiding major sample selection, agree­
costs, time, and effort to acquire the data. Secondary data will be data ment, and estimation errors through structural estimations with a vast
that has just been created for first-time use by the company and for later data set; and, in comparison to the World Bank database on fiscal con­
use by others. The researchers used the annual financial report of five straints, yielding progressive results for financial obstacles by utilising
savings and loan companies; it was sourced from the company’s website. financial statement data instead of survey data.
The study used trend analysis as a strategy. This method was used in the
study, in particular, to analyse trends in financial issues from 2014 to 3.2.2.2. Firm size. Three measurements of corporate size—total assets,
2018. In other words, the examination looked at the firms’ annual re­ total liabilities, and total equity—are expressed in natural logarithms.
ports. The implications for the creation and accessibility of the work Empirical results suggest a relationship between business size and in­
were extracted from the study. vestment strategy. (For examples, see Dang et al., 2018; Daunfeldt &
Hartwig, 2014; Graham & Harvey, 2001; Hartwig, 2012). Evidence
3.2.1. Model specifications suggests thatcompanies in nations with better-established financial
Firm size, corruption, and financial challenges of the selected firms

3
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

sector are increasingly dependent on external sources of capital, such as principal benefits of using panel data include accounting for the het­
debt and equity financing (Beck et al., 2002). erogeneity of each individual, greater flexibility, a reduced chance of
collinearity between variables, and more information in the data set. It is
3.2.2.3. Corruption perceptions index (CPI). Transparency International also easier to develop and test more complex models, and it is easier to
provides this index as a way to measure public sector corruption across examine the “dynamics of adjustment."
180 countries. The scale runs from 0 (corrupt) to 100 (clean) (Trans­ There are two types of panel data: balanced and unbalanced. If the
parency International, 2017). The best-known perception-based com­ data set needs to be balanced, cross-sectional and time-series observa­
posite index is Transparency International’s (TI) Corruption Perceptions tions must be collected. If there are the same number of observations in
Measure (CPM), which rates countries according to perceptions of cor­ the time series for each cross-sectional unit, the panel of data is said to be
ruption levels. It is a compilation of 13 other indices from 12 additional balanced.
organisations. The dataset is considered to be unbalanced since the time series
contains erroneous returns. Some companies’ recent stock market listing
3.3. collateral security is what accounts for the dataset’s gaps. Panel data estimates use one of
two types of effects models: fixed effects or random effects. Different
The phrase “collateral security” can be used to describe the insurance approaches might be used in the panel data regressions depending on
If, however, a borrower fails to fulfill their obligations as agreed, a the data. Depending on the components of the fixed effects model, a
specific asset gives a lender recourse. That is, if the borrower does not disturbance term (u!) is predicted to have various cross-sectional effects
pay the agreed-upon loan instalments, the lender may sell the collateral on the dependent variable (y!). The following are the definitions of the
to recoup all or a portion of the money advanced to the debtor. Ac­ terms:
cording to O’Sullivan and Sheffrin (2003), “collateral” is a guarantee u!" = μ! +v!" (3)
made by a borrower to a lender to provide certain property as security
for the repayment of a loan. If the borrower doesn’t adhere to the con­ where v!” is the time-varying disturbance term and μ! is the individual-
ditions of the loan agreement about principle and interest payments, the specific effect.
collateral shields the lender from a borrower default that might be used It is crucial to keep in mind that! doesn’t change throughout time.
to cancel the credit (Charya & Viswanathan, 2011; Biais et al., 2016; This model, sometimes referred to as the Least Squares Dummy Variable
Inderst & Mueller, 2007). Kitale et al. (2016) found that the collateral (LSDV) approach, is a relationship of several dummy-variables that
requirements of financial institutions had a significant influence on ac­ change based on the object.
cess to credit. This resulted in subpar performance since businesses
y!" = βx!" +μ!1+μ!D2+μ!D3+ … +μ!DN+υ!" (4)
without collateral were unable to get loans.
This equation may be used to determine whether pooled regression is
3.4. Private credit from banks appropriate for the dataset. For the first entity, the D1 is 1, and for each
successive dummy, it is 0. A fixed effects model for the time series is also
Non-bank lending that involves the issuance of debt but not its a choice (Brooks, 2014). The random effects model has separate in­
trading on the open markets is known as private credit. Direct lending or tercepts that are consistent throughout time for different entities. It is
private lending are other names for private credit. It falls under “alter­ important to note that with fixed effects, the intercept of each cross-
native credit.” One of the asset types with the highest growth has been sectional entity is affected by a separate intercept, which affects each
private credit (Flanagan, 2018). entity. The heterogeneity among the entities will be illustrated in this
example by a random term that varies cross-sectionally. The LSDV fixed
3.5. GDP effects model’s dummy variables, which reflect the heterogeneity, serve
as a representation of the cross-sectional element’s variability.
Growth in per capita income (PCI), a macroeconomic indicator, is The random effects in panel data are explained by the following
used to measure economic growth (Levine, 1997). This research uses equation:
real per capita GDP as a metric of growth, as is common in the literature.
y!" = α + βx!" + ω!" where ω!" = ε! + ν!" (5)
We adjust for income level since a more developed financial sector is
often seen in wealthy nations. Additionally, per capita income growth is Using either of the two models presented, endogeneity problems,
crucial since nations with rapid economic growth tend to have higher which are characterised by a link between the regressor and the error
credit demand and supply (Djankov et al., 2007; Emran & Farazi, 2009). term, may be addressed (Brooks, 2014). The parameter estimate and
inference will be skewed whenever the regression has endogeneity is­
3.6. Cash holding sues (Roberts and Whited, 2012).

“Cash holdings,” are the amount of cash and cash equivalents, highly 4. Results and discussions
liquid assets that a corporation may swiftly convert into cash (Hassa­
nein, 2021). 4.1. Descriptive statistics of the Analysis’s sample

3.7. Exchange rate In this section of the study, the descriptive analysis of the selected
variables are analyzed. Some descriptive statistics include the mean,
The cost of converting two distinct currencies between two different median, maximum, minimum, standard deviation, asymmetry, kurtosis,
nations or economic areas is known as the exchange rate of a currency. It sum, squared difference, and different observations. Table 1 provides a
is used to determine how much one currency is worth in relation to good illustration of these numbers.
another and is essential for estimating the movement of money and trade The variables have a positive mean and median, as shown in Table 1.
trends (CorporateFinanceInstitute.com). Given the model’s variable inclusion, this is expected. Given that the
variables’ modest standard deviations are thought to indicate only
3.7.1. Panel data moderate variation of these variables throughout the course of the
Using panels to organise data enables testing and analysis of more research, for each of the model’s variables, the maximum and minimum
complex datasets (Brooks, 2014). Baltagi (2005) indicated that the values fall between 1.6 and 0.009, respectively. Study variables,

4
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

Table 1
Descriptive Statistics of the variables.
Financial challenges Collateral Cash holding Firm size Corruption index GDP Private credit Exchange rate

Mean − 0.570947 0.054536 0.112042 8.216435 44.16667 4.755061 34.16915 3.787883

Median − 0.448266 0.053175 0.112136 8.237311 44.5000 4.523896 34.60376 3.997800
Maximum − 0.075221 0.103340 0.168284 8.374579 48.0000 8.143447 38.20729 4.83000
Minimum − 1.729414 0.008634 0.066914 8.001132 40.0000 2.178207 30.55150 2.161600
Std. Dev. 0.612678 0.046573 0.035562 0.142098 3.311596 2.292173 2.874416 0.981106
Skewness − 1.247794 0.017185 0.337776 − 0.395328 − 0.123334 0.312590 − 0.017801 − 0.667313
Kurtosis 3.265986 1.049428 2.222073 1.824851 1.387875 1.707717 1.825525 2.238757
Jarque-Bera 1.574677 3.503722 0.265385 0.501528 0.664948 0.515212 0.345165 0.590180
Probability 0.455054 0.173451 0.875734 0.778206 0.717147 0.772900 0.841489 0.744465
Sum − 3.425684 0.327217 0.672255 49.29861 265.0000 28.53037 205.0149 22.72730
Sum Sq. Dev 1.8768 0.010845 0.006323 0.100959 54.83333 26.27030 41.31134 4.812846
Observation 60 60 60 60 60 60 60 6

Source: Author’s Construct, 2022

including financial constraints and the log of assets, are negatively Lehman et al. (1998) separately suggest 0.9 and 0.7 as edges of bivariate
skewed in terms of skewness, whereas collateral and cash hold are connections. The correlation matrix’s results reveal that the indepen­
positively skewed. dent variables don’t exhibit any multicollinearity when the edges are
The Jarque-Bera measurement that demonstrates the null hypothesis taken into consideration.
that all the series are gotten from a normally distributed random process
cannot be rejected for financial constraints (total asset, collateral, cash 4.1.2. The level of financial challenges of firms
hold), implying that they are normally distributed. As a result, ordinary The Whited-Wu index was utilised in this section to determine the
regression analysis was used to determine the determinants of the company’s level of financial challenges. The analysis is summarized
financial challenges faced by Ghanaian savings and loan companies. Table 3 and Fig. 1.
Table 1 also shows that financial challenges, total assets, collateral, Table 3 illustrates the log total assets (LTA), cash flow (CF), dividend
and cash holdings make it impossible to reject the Jarque-Bera mea­ payments (DIVP), total long debt to equity (TLDD), firm sales growth
surement, which shows the null hypothesis that all the series are ob­ (FSG), and industrial sales growth (ISG) that were used to determine the
tained through a normally distributed random process. This suggests level of financial challenges of the firms. In 2016, the cash flow-to-total-
that the series are normally distributed. In order to identify the factors assets ratio was 251.6, and the long-term debt-to-total-assets ratio was
that contribute to the financial difficulties experienced by Ghanaian 0.21, with a growth rate of 54.1% (the industrial growth rate was 5.1%).
savings and loan organisations, conventional regression analysis was This translated into a higher level of financial challenges of − 1.7 from
performed. 2016 to 2017, meaning the company had easier access to funds
compared to the period between 2017 and 2018 (Fig. 1). The Bank of
4.1.1. Correlation matrix Ghana’s financial sector reorganisation restricted enterprises’ access to
The correlation matrix is displayed in Table 2. The highest correla­ capital in 2021 (− 0.33667), raising the minimum amount of capital that
tion (0.449) between the two variables, financial challenges and firm financial firms are required to have.
size, can be seen, demonstrating the strong relationship that exists be­
tween the two. It is acknowledged that the high correlation does not 4.1.2.1. Panel unit root. Despite the fact that a dataset covering 6 years
pose a problem for this model because the link is positive and that it is is not especially long, it is very probable that the economic indicators
reasonable to anticipate similar relationships between the independent will possess a unit root (Nelson and Plosser, 1982). We apply the panel
factors and the dependent variable given the high correlation. Since no unit root test to determine the pattern of integration for each of the
two variables were substantially linked, multicollinearity was not a series in our sample. The study used the Fisher-type panel unit root test
problem in these estimations, according to coefficients among most of since the dataset was unbalanced. If the variables to be examined are I
the variables. (0) and I(1), it is not required to examine the order of the variables’
As shown in Table 2, cash holdings have a negative correlation with integration when using the dynamic fixed effect, mean-group, or pooled
financial challenges and the corruption index, while private credit from mean-group estimators (Pesaran and Smith, 1995; Pesaran, 1997;
banks and GDP have positive correlations with financial challenges. Pesaran et al., 1999). No series should exceed the I (1) order of
Finding multicollinearity between the variables is the main goal of the magnitude; however, we do these tests.
connection between the variables. Collinearity, according to econo­ Table 4 shows the results of the unit root tests, which show that one
metrics references, raises parameter difference evaluations, produces of the variables under investigation (GDP) is stationary in order I(0),
high R-squares despite low parameter hugeness, and causes parameters whereas firm size, collateral security, cash holdings, the corruption
to have incorrect signs and impermissible extents (Mela and Kopalle, index, and private loans from banks are integrated in order I(1). The
2002). For the negative impact of collinearity, Green et al. (1988) and results show that all variables are stationary at the first difference

Table 2
Correlation matrix for the variables.
Financial challenges Collateral Cash holding Firm size Corruption index GDP Private credit Exchange rate

Financial challenges 1.000

Firm size 0.4490 1.000
Collateral 0.2535 0.2402 1.000
Cash holding − 0.0592 0.4705 0.2458 1.000
Corruption index 0.4004 0.4094 0.2266 0.483 1.000
Private credit from banks 0.3207 0.2751 0.1947 0.4162 0.4800 1.000
GDP 0.3037 0.1920 0.1613 0.3945 0.4162 0.4654 1.0000
Exchange rate 0.0846 0.0583 0.1884 0.1649 0.1332 − 0.1967 0.1714 1.0000

5
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

Table 3
The Level of Financial Challenges of the firms.
Year Log total asset Cash flow Dividend payment Long term debt to asset Firm sales growth Industrial sale growth Financial constraints

2016 8.001132 251.5894 0 0.209546 54.14129 5.1 − 1.72941

2017 8.113798 248.2191 0 0.194963 20.44207 5.1 − 0.55987
2018 8.191198 245.9958 0 0.744128 24.57972 5.8 − 0.63334
2019 8.283425 243.3776 0 0.751532 11.67592 6.8 − 0.09118
2020 8.334481 242.0067 0 0.064938 5.718913 5 − 0.07522
2021 8.374579 240.9673 0 0.083386 10.82736 4.2 − 0.33667

Fig. 1. The level of financial challenges and firm growth rate.

Table 4 Table 5
Results of the panel unit root tests. Results of the Westerlund (2007) panel cointegration test.
Fisher-type test Financial Challenges

Levels 1st Difference Value Z-score P-value

Variables Test values p-value Test value p-value Gt − 4.085 − 11.769 0.000
Financial challenges − 0.976 0.08 − 14.6847 (0.00)*** Ga − 14526 − 1.318 0.000
Firm size − 0.8174 0.1734 − 12.38 (0.000)** Pt − 17.003 − 7.878 0.225
Collateral − 0.7800 0.1816 − 13.8457 (0.00)*** Pa − 17.608 − 7.882 0.000
Cash holding 1.2812 0.0992 − 12.364 (0.000)**
Corruption index − 0.6456 0.2612 − 15.0936 (0.00)*** The Westerlund (2007) tests consider the absence of cointegration to be the null.
Private credit from banks − 1.2536 0.0902 − 17.93 (0.00)*** A constant and a range of lags (1–2) and leads are used to fit the test regression (1
GDP − 5.976 0.00 − 14.6847 (0.00)*** 2).
Exchange rate − 1.1740 0.2116 − 16.36 (0.000)** Source: Author’s constructs, 2022

average absolute value of the off-diagonal components is 0.381. This

(Amankwaah et al., 2022).
blatantly implies that the cross-sectional units are affected by cross-
sectional dependence or a common cause.
4.1.2.2. Panel Co-integration. The cointegration test shows if the inte­
For the test statistics’ dependable critical values, we bootstrapped.
grated variables have any long-term associations. Too short delays may
Bootstrapping decreases the discrepancy between the actual and nomi­
not adequately represent the dynamics of the 55 system, causing vari­
nal rejection probability and corrects crucial values in tests. Our findings
ables to be excluded, biassing the remaining coefficients, and even
show that for the entire sample, the null of no cointegration is acceptable
resulting in serially linked errors. A latency that is too lengthy, however,
in PT test statistics, whereas GT test statistics accept the null of no
causes over-parameterization and a quick loss of degrees of freedom
cointegration for financial challenges. However, when Pt is constrained
(Kireyev, 2000).
to be homogeneous, at the 5% level for Gt and the 10% level for Pt, the
The results of the Westerlund (2007) cointegration test are summa­
null hypothesis of no cointegration is rejected, indicating that the entire
rized in Table 5. The cointegration test reveals that while firm size,
panel is cointegrated for financial challenges. The outcomes using
corruption, and financial challenges are not correlated, they are corre­
bootstrapped p-values offer more convincing proof of cointegration. The
lated with each other. However, when examining the possibility of a
computed asymptotic and bootstrapped p-values are shown in Table 6.
long-term connection between firm size, corruption, and financial dif­
ficulties, only Pt test statistics accept the null of no cointegration. There
is therefore some evidence of cointegration. 4.2. Hausman Test (fixed effect/random effect)
The results of the cross-sectional independence Pesaran test, which
show a significant absolute correlation, are 39.857, Pr = 0.0000, and the To choose between the fixed effect and the random effect, the

6
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

Table 6 Table 8
Results of the Westerlund (2007) panel cointegration test, bootstrapped. Results of the Long-Run and Short run Coefficient.
Financial Challenges Financial Challenges

Value Z-score P-value Robust P-value Variables Coefficient Std. Err. Z-score P>|t|

Gt − 3.202 − 0.847 0.039 0.025 Long run Coefficients rowhead

Ga − 8.912 − 3.948 0.049 0.031 Intercept − 2.945085 1.16461 − 2.53 0.020
Pt − 10.374 − 9.573 0.042 0.026 Firm size − 9.516842 1.2265 − 7.76 0.000
Pa − 9.284 − 1.264 0.063 0.057 Collateral − 11.35841 2.4568 − 4.62 0.000
Cash holding 10.69209 2.0637 5.18 0.000
Source: Author’s constructs, 2022 Corruption index 0.216087 0.0562 3.84 0.001
Private credit from banks 0.2856639 0.0511 5.59 0.000
researchers used the Hausman Test. The standard is that if the Chi- GDP − 0.0040695 0.0283567 − 0.14 0.887
Exchange rate 0.5745877 0.0506871 − 2.53 0.020
square measurement acquired by the Hausman Test is bigger than the
basic Chi-square X2 = 21.03, then the random effect estimator is the
fitting model for the study. The outcomes showed a Chi-square value of
Short run coefficients rowhead
86.18 for financial challenges. This is more prominent than the Chi-
square basic worth at the 5% significant level. This means that the dis­ Error Correction Coefficient − 0.104 0.022 − 6.68 0.000
tribution of the co-efficient is ordered and gives proof for the random Firm size − 0.2643 0.071 − 3.70 0.001
Collateral − 0.060 0.025 − 2.41 0.021
effect model. The result is recorded in Table 7. Cash holding 0.006 0.005 2.16 0.037
Corruption index 0.012 0.010 − 2.38 0.022
4.2.1. Impact of firm size and corruption on financial constraints Private credit from banks − 0.375 0.082 − 5.094 0.000
This section analyse the effect of corruption index, firm size on GDP 0.018 0.0103 1.75 0.088
Exchange rate − 0.07 0.0561 − 7.882 0.000
financial constraints of the firms using panel analysis.
Intercept 0.5169 0.5096 − 1.01 0.316
According to Table 8, there is a negative and statistically significant
correlation between total assets and financial difficulties. All other
things being equal, a rise in total assets will result in a general decline in inefficiently. The action could result in the loss of clients who lose faith
financial difficulties. More significantly, a change in the total assets as a in the company and turn to competing products, causing disasters. In
percentage will result in a 95.2% reduction in financial difficulties. This addition, internal or external corruption may drive an organisation to
means that as a company grows, its ability to pay off debt becomes less inflate its costs to recoup lost assets. Competing firms may take advan­
problematic. tage of this fortunate break to outperform the influenced firm, resulting
According to Table 8, there is a link between financial issues and the in a significant loss of market share and, as a result, financial difficulties.
collateral needed by the business to obtain a loan that is both negative The degree of financial imperatives for business and nation development
and statistically significant. All other things being equal, a rise in appears to be an endless loop: a terrible business condition will increase
collateral will result in a general decline in financial challenges. A per­ financial challenges for organisations, and heavier financial challenges
centage adjustment in the collateral will lessen financial difficulties by will negatively affect the economic growth and development of the
113.58 percent, which is more significant. This could be due to firms’ nation because of lower business performance. Desai et al. (2003) noted
lack of good collateral for obtaining external funds to support their ac­ that capital limitations because of institutional defects affect business
tivities. Collateral constraints Because of the downturn in the financial passage and development limits, particularly in less developed nations.
market, a reduction in resource costs brings about a lower estimation of This conforms to Chan’s (2009) findings that defilement can have a
the assurance and, accordingly, a lower estimation of the credit that can genuinely negative effect on the development of organisations con­
be obtained with this guarantee. fronting financial challenges.
The financial difficulties of the company are positively and statisti­ Private credit and the financial difficulties of the enterprises are
cally significantly correlated with the bank’s cash holdings (Table 8). statistically significantly correlated in Table 8. This suggests that a
More crucially, with everything else being equal, a change in the firm’s 28.6% increase in financial issues followed a percentage gain in finan­
cash holdings will lead to an increase in financial difficulties of 113.6%. cial progress. This, however, confirms a prior prediction that financial
The implication is that holding more cash will lead to a low level of fixed development has a positive effect on the firm’s financial challenges.
assets and, hence, a reduction in the net return on fixed assets, which Financial development has the potential to transform firms’ existing
depend on capital turnover. financial activities to meet the current needs of customers, which may
With respect to the corruption index (a measure of institutional result in financial challenges. If the available funds for funding such
development), statistics has determined that the calculated coefficients activities are not there, Beck et al. (2008) show that since independent
in Table 8 are significant (P < 0.05). The corruption index is positive and companies think it’s increasingly hard to get to financial services
statistically significant in relation to financial challenges. This implies because of higher information and exchange costs, financial develop­
that, all other things being equal, the percentage increase in corruption ment will help them lopsidedly. Small organisations are more averse to
reduces financial challenges by about 21.6%. This confirms a prior acquiring money at market interest rates and, consequently, are liable to
prediction of a negative relationship. Inferentially, corruption credit proportioning (Audretsch & Elston, 2002). Love (2003) found that
frequently results in the loss of assets through theft and graft. Assets that financial development remained critical in the wake of controlling the
could be used to carry out corporate strategies either fail or are used size of the company, monetary cycles, and the development of the
legitimate framework. Similar findings were made by Beck and
Demirguc-Kunt (2006), who found that the degree of property rights
Table 7 significantly influences whether or not small and large businesses can
Results of the Hausman test. close the external finance gap.
Financial Challenges Further, findings from the table show that the economic develop­
ment (measured by GDP) of Ghana is statistically insignificant compared
H0: Random Effect Model
H1: Fixed Effect Model to the financial challenges of the firm. This means that the percentage
Chisq 86.18 increase in economic development does not influence the financial
P-value 0.0001 challenges of the firms. This clearly indicates that economic

7
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

development cannot pose any challenges to retaining skilled employees has recommended new guidelines for the liquidity coverage ratio (LCR)
in the firm, which can result in upward pressure on wages, hence no and the supplemental leverage ratio (SLR), which should be adhered to.
financial challenges for the firms. This supports Gorodnichenko & The research advises savings and loan firms to restructure their
Schnitzer’s (2011) findings that financial challenges restrict the capacity commercial operations. Company realignment is mostly done to get out
of national organisations to advance and, in this manner, arrive at of lines of business with poor margins and go into lines that are more
progressively developed countries. They claim that institutions in less inherently fiscally sound and boost bank profitability. Leading banks use
industrialised nations face financial difficulties and won’t have adequate a robust planning approach, examining the base cost of assets antici­
funds to put resources into imaginative exercises that may keep them pated to compete in a certain industry and looking for opportunities to
from receiving better advancements. For instance, Love (2003) shows differentiate themselves from rivals. This suggests that S&L corporations
that monetary cycles can likewise influence business financing. may elect to enter atypical organisations in a number of circumstances,
The negative relationship between exchange rate and financial such as by investing in other ventures that might generate steady rev­
challenges shows that the exchange rate significantly influences the enue. This might aid the business in successfully and efficiently
financial challenges of a firm. Private investments have greater ex­ competing with other players in the financial sector.
change costs and risk premiums due to fixed exchange charges and in­ The study also recommends firm strategies for channel optimization.
formation asymmetries since they are less clear-cut, have fewer promises Channel optimization is an important part of bank advancement, with
to make, and generally don’t have the same risk. examined financial branches being closed, merged, sold, and bought. The personnel, func­
statements that may provide a more complete view of the firm and its tions, and responsibilities of banks’ branches are being reorganized, and
expected profit (Beck, 2007). Higher advances are a result of fixed ex­ new criteria are being used to evaluate the effectiveness and worth of
change costs related to credit evaluation and handling. each branch.
There is a need to improve market share to enhance performance and
4.2.2. Model evaluations and stability checks reduce financial challenges. This can be done through business channel
The assessed characteristics of a time series may alter with time, optimization solutions. The goal of channel improvement is to assess the
according to Hansen (1992). To avoid model misspecification that might various customer communication channels and create a mix of financial
occur from unstable parameters and, as a result, cause a bias estimate, it literacy that is tailored to each bank’s unique clientele.
is imperative to do parameter testing (Amankwaah et al., 2022). Table 9
displays the model diagnostics and goodness of fit findings. 5.2. Limitations
Table 9 demonstrates that the errors are evenly spaced out, then the
model is valid both the white heteroskedasticity test and Ramsey’s The results of this investigation depend generally on the analysis of
RESET for accurate model specification (Amankwaah et al., 2022). secondary data. Subsequently, the results of the investigation are
dependent upon the impediments in the bank’s financial statement as
5. Conclusion unveiled to the public, which was under the supervision of the Bank of
Ghana. The available information for the study periods of 2013 and
All things considered, the firm’s operations and profitability are 2018 was difficult to come by. The management of Savings and Loans
greatly impacted by a lack of adequate capital. There is a strong corre­ Company felt reluctant to provide their annual report, making it difficult
lation between the firm’s operations and enough finance. This just im­ to draw generalised conclusions.
plies that the accessibility of satisfactory financing helps support the
operations of savings and loan companies. This record infers that there Ethics approval and consent to participate
should be adjusted prudential guidelines for savings and loans at all
levels in Ghana, and this arrangement is occasionally explored. The The manuscript has been consented and ethically approved by the
study noted that total assets, collateral, and cash holdings have a big authors.
impact on how much money savings and loan firms may borrow. The
researchers noted that issues identified with the minimum capital Consent for publication
requirement, capital adequacy, liquidity, and reporting requirements
should be less friendly for savings and loan companies. As a result, su­ The authors agreed to publish the manuscript.
pervision and monitoring became less viable and were identified as a
significant reason for the company’s expanding financial challenges. CREDIT AUTHOR STATEMENT

5.1. Policy recommendations Eric Amankwaah: Investigation, Formal analysis, Writing – original
draft, Data curation, Writing – review & editing. Nana Baidoo: term,
Savings and Loan Companies should review their deposits and asset Conceptualization, Methodology and Resources
mix on an annual basis to ensure that they are in compliance with reg­
ulations while also not settling for increased profitability. Banks should Declaration of competing interest
intentionally expand cross-selling activities and enhance customer
relationship management in order to keep deposits. The Bank of Ghana 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.
Table 9
Model goodness of fit and model diagnostic tests.
Availability of data and materials
Test Statistics LM Version F Version

A: Serial Correlation *CHSQ(1) = .40044(.841) F(1, 59) .30539 The datasets used during the current study are available from the
[.862]
corresponding author on reasonable request.
B: Functional Form *CHSQ(1) = .26429(.607) F(1, 59) .20263
[.656]
C: Normality *CHSQ(2) = 29.0231 Not applicable Abbreviation
(.000)
*D: Heteroscedasticity*CHSQ(1) = .27887[.597] F(1, 40) .26737 BoG Bank of Ghana
[.608]
S&LC Savings and Loans Companies

8
E. Amankwaah and N.O. Baidoo Social Sciences & Humanities Open 8 (2023) 100552

UNDP Human Development Report Daunfeldt, S. O., & Hartwig, F. (2014). What determines the use of capital budgeting
methods? Evidence from Swedish listed companies. Journal of Finance and Economics,
SDI Specialized Deposit-Taking Institution
2(4), 101–112.
PNDC Provisional National Defence Council Desai, M., Gompers, P., & Lerner, L. (2003). Institutions, capital constraints and
GHASALC Ghana Association of Savings and Loans Companies entrepreneurial firm dynamics: Evidence from europe. In NBER working paper series,
LSDV Least Squares Dummy Variable working paper 10165.
Graham, J. R., & Harvey, C. R. (2001). The theory and practice of corporate finance:
Evidence from the field. Journal of Financial Economics, 60(2), 187–243.
References Hartwig, F. (2012). Four papers on top management’ s capital budgeting and accounting
choices in practice. Uppsala University.
Abuhommous, A. (2016). Financial constraints, capital structure and dividend policy: Hassanein, A. (2021). Corporate governance and cash holdings. In A. Alqatan,
Evidence from Jordan. Brunel University. K. Hussainey, & H. Khlif (Eds.), Corporate Governance and its Implications on
Amankwaah, E., Mensah, N., & Baidoo, N. O. (2022). The impact of location tax Accounting and finance (Eds., pp. 305–321). https://doi.org/10.4018/978-1-7998-
incentives on the growth of rural economy: Evidence from Ghana. Future Bus J, 8(1), 4852-3.ch014 IGI Global https://www.investopedia.com/terms/c/collateral.asp
54. https://doi.org/10.1186/s43093-022-00163-3, 2022. https://www.unodc.org/e4j/en/anti-corruption/module-1/key-issues/measuring
Beck, T., Demirguc-Kunt, A., & Maksimovic, V. (2002). Financial and legal constraints to -corruption.html https://corporatefinanceinstitute.com/resources/knowledge/eco
firm growth : Does size matter?. In Policy research working paper;No. 2784. nomics/exchange-rate/.
Washington, DC: World Bank. © World Bank https://openknowledge.worldbank. Levine, R. (1997). Financial development and economic growth: Views and agenda.
org/handle/10986/15616 License: CC BY 3.0 IGO. Journal of Economic Literature, 35, 688–726.
Dang, C., Li, Z. F., & Yang, C. (2018). Measuring firm size in empirical corporate finance. O’Sullivan, A., & Sheffrin, S. M. (2003). Economics: Principles in action. Upper saddle river,
Journal of Banking & Finance, 86, 159–176. New Jersey 07458. 513. Pearson Prentice Hall. ISBN 0-13-063085-3.