Journal of Entrepreneurship and Business Venturing

Vol 4, Issue 1, 2024, PP. 217-240

Venturing into Risk Management: Examining Credit Risk and Performance in

Pakistani Commercial Banks through Panel VAR Analysis from a Business
Perspective

Raza Ali
Faculty of Management Sciences, SZABIST University Karachi, Pakistan.
Jameel Ahmed
Faculty of Management Sciences, SZABIST University Karachi, Pakistan.
Financial Stability Department, State Bank of Pakistan, Karachi, Pakistan.
jameel.khilji@gmail.com

Corresponding: raza.soomro@hotmail.com
ARTICLE INFO ABSTRACT
Article History: In Pakistan's dynamic financial sector, effective credit risk management by
Received: 13 Oct, 2023 commercial banks is critical to maintaining financial stability. In this study, we
Revised: 25 Dec, 2023
Accepted: 20 Feb, 2024 embarked on a unique and comprehensive journey, adopting the Panel Vector
Available Online: 29 Feb, 2024 Autoregressive (PVAR) analysis, a novel methodology in this context. This approach
allowed us to delve into the intricate relationship between credit risk management
strategies and the aggregate business performance of commercial banks. The data
DOI:
https://doi.org/10.56536/jebv.v4i1.101 used in this study spans from 2010Q1 to 2022Q4, investigating the comprehensive
nexus of credit risk and macroeconomic indicators using a large dataset encompassing
18 strategically critical commercial banks of Pakistan. Our study uncovers several
Keywords:
Credit Risk Management, Panel profound insights, documenting a remarkable increase in credit risk upon increases in
VAR Methodology, Non- the interest rate. The repercussions of the non-performing loans are observed nearly
performing Loans, Economic two years after the successive phases of a stern posture of monetary policy have been
Growth, Inflationary Trends,
Commercial Banking,
implemented. Our research findings carry significant weight for the banking industry.
Macroeconomic Dynamics, We discovered that the levels of credit risk could be reined in due to the protective
Pakistan. shield offered by phases of sudden economic booms. This underscores the reinforcing
effect of the improving economic conditions on defaults of loans, providing practical
JEL Classification:
G20, G21 and actionable insights for credit risk management strategies.
© 2024 The authors, under a Creative Commons Attribution-Non-Commercial No-Derivatives 4.0.

INTRODUCTION

It is unsurprising that credit risk management has been described as fundamental to the success
and stability of the banking systems worldwide. This seems particularly true in the complex money
calculus of the developing world; however, it calls for understanding global academic research and
pragmatic governance (Harb et al., 2022). One of the most critical activities in the banking industry is
credit risk management (CRM). A bank takes risks to make a profit. As such, proper CRM also enables
a bank to improve its overall performance and to survive in times of financial downturn. An effective
CRM can minimize a bank's loan defaults and non-performing assets, which can significantly
undermine the stability and soundness of a bank’s loan portfolio. A well-structured CRM framework
also gives the bank's management team the tools to underwrite loans soundly, allocate capital
efficiently, and develop and maintain strong borrower relationships (Mennawi, 2020).

217
Ali & Ahmed JEBV 4(1) 2024, 217-240

Credit risk, also referred to as default risk, is the probable risk a bank assumes when it makes a loan,
line of credit, or any other form of financing to a borrower. Banks deploy a powerful credit risk
management system to identify and manage pair risk through this. This means they work on their
ability to detect, understand, and monitor credit risks to avoid facing default. Therefore, investing in
robust credit risk management to increase the likelihood of default avoidance is crucial for banks to
successfully and safely maintain their financial position (Rakhaev, 2020). It is well-supported by
research, and many studies have unfolded on how effective credit risk management is a function of
bank performance (Bhatt et al., 2023).

A literature review from a Pakistani vantage point identified several salient conclusions regarding
credit risk management variables, key bank performance measures, and macro-level indicators from
earlier research literature and the constructed frameworks of variables and their relationships, such as
Zubair and Sajid (2014). They emphasized the continuing relevance of these interrelationships,
particularly highlighting the strong negative relationship between one of the most critical bank
performance measures of Gross NPL Ratio and key bank performance measures such as Return on
Assets and Equity. The authors suggested that some of the specific phases of the management of credit
risks, such as identification analysis and valuation, do not have a direct, substantial impact on financial
robustness and strength as this is positioned with profit and loss strategies and interpretations, a
perhaps striking exception to the general rule or direction in the literature revealed by Lafayette,
Vaillant, and Vendrell-Herrero (2019).

With its peculiar economic evolution, regulatory landscape, and market exigencies, the Pakistani
banking sector is an ideal context to scrutinize this paradigm. Our study seeks to reveal how credit risk
management is related to the performance of commercial banks in Pakistan. The banking sector in
Pakistan shares overlapping traits with other developing economies yet possesses its challenges and
prospects. How its banks perform on the ground relative to micro and macroeconomic factors and
navigate the ongoing progress between their regulatory environment and operating conditions provides
an arena of splendid scholarly pertinence.

The panel Vector Autoregressive (VAR) methodology chosen for our reintegrating analysis is not
accidental, given that VAR methodology enables a thorough analysis of multivariate time series data,
as it captures the interdependent behavior among endogenous variables over time. Using our chosen
methodology, we appraised performance dynamics within the sector, tapping both bank-specific
metrics and acknowledging broader macroeconomic indicators. The set of variables chosen has been
thoroughly selected based on existing literature and our research objectives to provide an extensive
outlook of the sector characteristics. The chosen variables constitute a wide range of financial
indicators, including profitability indices and liquidity indicators.

218
Ali & Ahmed JEBV 4(1) 2024, 217-240

The motivation behind this academic endeavor is twofold: (a) to enhance the understanding regarding
the symbiotic relationship between commercial banks’ vitality and credit risk management under the
canopy of an emerging economy and (b) to furnish scholars, practitioners, and policymakers with the
valuable insights, which may allow them to synchronize the banking praxes with the tenets extracted
from the pool of academicians’ celebration.

Our study examined the performance of 18 major banks in Pakistan from 2010 to 2022 to determine
their risk management strategies. We have collected data from official bank repositories, annual
reports of banks, and regulatory bodies. In this tedious journey, we faced several challenges, including
data consistency, heterogeneity in reporting standards, and evolving regulatory requirements. Bank
profitability, capital adequacy, and macroeconomic factors were gauged. We found that interest rates
significantly affect the riskiness of lending for banks, most likely due to the increased risk of loan
defaults when lending becomes riskier as interest rates rise. Our results suggest that in addition to
bank-specific variables such as profitability and solvency, credit risk is also significantly influenced
by the monetary policy stance, economic activity, and inflation. While the reaction of credit risk to
shocks to micro-level idiosyncratic variables is immediate, shocks to macro-level systemic variables
manifest their impact after a lag. These findings have policy implications for the banks as well as
regulators.

The structure of the paper is as follows. After this introduction, the next section is devoted to a
literature review that sets the context and rationale for the research. Section 3 outlines our data
methodology and discusses the techniques and datasets used. Section 4 presents the results and the
discussion, which analyzes and interprets our findings on how various economic factors are related to
credit risk in Pakistan's banking sector. Section 5 concludes our study, summarizes the main insights
gained, and underscores their implications for policymakers and practitioners in the banking industry
who seek to develop a comprehensive understanding of the dynamics of credit risk management in
Pakistani commercial banks and formulate strategies for better and more sustainable management of
financial stability.

LITERATURE REVIEW

This research aims to identify the relationship between credit risk management, bank
performance, and macroeconomic variables. The second section of the literature review discusses the
relationship and figures between Gross Non-Performing Loan Ratio (GNPLR), Return on Assets
(ROA), Capital Adequacy Ratio (CAR), Inflation rate, GDP, and Interest rate. This literature review
also contains the summarized findings of the above variables. Besides, the last part of the literature
review critically analyzes the significant findings and suggests the implications for Pakistan's
economy.

219
Ali & Ahmed JEBV 4(1) 2024, 217-240

Adverse macroeconomic and bank-specific conditions lead to elevated credit risk in the banking
industry. Recently, Alazis (2020) revealed that the Gross Non-Performing Loan Ratio (GNPLR) has
a negative association with Return on Assets (ROA) and Capital Adequacy Ratio (CAR), which are
the core performance indicators of the banks. Similarly, a recent study by Khurram and Siddiqui
(2021) also found that inflation and GDP growth have a positive relationship with credit risk.
However, the interest rate does not significantly affect credit risk.

Macroeconomic and bank-specific adversities magnify credit risk in the banking industry. Recently
Alazis (2020) unveiled that Gross Non-Performing Loan Ratio (GNPLR) is negatively associated with
Return on Assets (ROA) and Capital Adequacy Ratio (CAR), which are the core performance
indicants of the banks. In addition, a recent study by Khurram and Siddiqui (2021) also found that
inflation and GDP growth have a positive association with credit risk. However, interest rates do not
significantly affect credit risk. The findings propose that effective credit risk management and stability
of macroeconomic variables are pivotal to stimulating growth and stability in Pakistan's banking
sector.

Banks’ adoption of Islamic finance in credit risk management constitutes a nascent nexus for
promoting entrepreneurship. In the absence of Islamic finance, start-ups struggle to obtain funds
through conventional finance since the high cost of debt undermines their potential. In contrast,
Islamic finance is a participative and equitable alternative to traditional finance by sharing objectives
and risks with entrepreneurs. The absence of a supportive regulatory environment, however,
encourages Islamic financial institutions to adopt a costly compliance approach, which has increased
compliance costs for Islamic banks and thus pass on higher costs to their entrepreneurs, particularly
amid high credit default rates resulting from non-performing loans (Arshed et al., 2023).

According to the research by Nuta et al. (2024) on the market performance of Romanian firms during
the COVID-19 crisis, it is possible to draw a perspective on credit risk management in banks. The
authors reveal significant differences in market performance by sector, with some sectors experiencing
accelerated growth during the crisis. These differences imply that banks lending to firms in different
sectors will have different levels of credit risk exposure. For example, communication services grew
during the period, suggesting that banks lending to firms in this sector had lower credit risk than those
in industries experiencing poor performance. Additionally, this study reveals the resilience and rapid
recovery of the stock market, which quickly rebounded from the initial shock of the pandemic,
suggesting that while short-term credit risks.

Gross Non-Performing Loans Ratio (GNPLR) and Bank Performance:

Gross Non-Performing Loan Ratio (GNPLR) is the most critical measure of a bank's asset quality,
which allows an investor to assess the overall health of a bank's lending portfolio and the extent of
risk associated with a bank's lending. A higher GNPLR indicates deteriorating asset quality and
potential weakness in credit risk management, signaling the need for robust risk mitigation strategies
220
Ali & Ahmed JEBV 4(1) 2024, 217-240

(Ponomarenko et al., 2017). Effective credit risk management can curtail the losses arising from non-
performing loans. Banks can shield their finances by deploying stringent measures to soften their
exposure to delinquent borrowers.

Research has confirmed an inverse correlation between the non-performing loan ratios (GNPLRs) and
principal performance indicators of banks, such as return on assets and equity (Dhurup et al., 2022).
They revealed the extent to which heightened GNPLRs affect a bank’s profitability and financial
soundness. Consequently, several scores fall well below credit risk management as banks identify,
analyze, and evaluate credit risks. Placing the chances of default under given lending risks, ways of
mitigating these risks, and providing resources minimizing these risks are allocated. This allows banks
to write loans more confidently after identifying and evaluating potential credit risks and monitoring
borrowers' worthiness. Thorough credit risk management is thus vital to minimize their GNPLRs and
potential (Barjaktarović et al., 2022).

Effective credit risk management empowers banks to identify, assess, and proactively mitigate
emerging risks, thereby enabling banks to build a strong loan portfolio and ensuring long-term
sustainability and profitability for the lending institution (Spuchlakova & Misankova, 2017). Banks
are well-advised to throttle back on loans that are not paying when due. The International Monetary
Fund reports that banks with higher levels of non-performing loans are much less profitable than those
with lower delinquent loans and are weaker during economic distress. Banks must act to manage and
lessen the level of non-accruing loans to maximize the bank's bottom line over the long haul (Ando,
2019).

According to Safar et al. (2021), the Gross Non-Performing Loan Ratio (GNPLR) enables banks to
assess the efficiency of their credit risk management. Closely monitoring GNPLR helps banks
recognize their weaknesses and take timely actions to maintain their financial health and profitability.

Credit risk management plays a critical role in helping to keep a bank financially stable by reducing
the impact of non-performing loans (Shrestha, 2017). A bank will be more likely to maintain its market
of non-performing loans (Arora, 2021), Allowing a bank to keep a sustainable financial performance
and adjust quickly not to become bankrupt and stay fully functional while doing this (Ibiz et al., 2021).
This study concludes that the main feature of credit risk management's impact on banks' financial
stability is to have a capital adequacy ratio and liquidity as they help investors measure the degree of
risk involved in buying particular investments (Evbayiro & Osagie, 2023).

Banks struggle to address non-performing loans and defaulters employing conventional strategies.
Indicators such as the Non-Performing Loan Ratio (NPLR) and the Capital Adequacy Ratio (CAR)
are keenly examined by banks due to their significance on the soundness, profitability, and risk
management by banks (Rehman et al., 2019). Key stakeholders are concerned about low non-
performing loans, transparent risk management, and balanced interests (Gunningham, 2020), as the
former indicates that borrowers are defaulting and increasingly placing financial risks. Minimizing
221
Ali & Ahmed JEBV 4(1) 2024, 217-240

NPLR, like Expected Loss, lowers credit cycle exposure and maximizes banks' profits (Erdoğdu,
2017).

GNPLR is consistently associated with various bank performance measures. A study on Pakistani
banks found a solid and significant negative association of GNPLR with both Return on Assets (ROA)
and Return on Equity (ROE). It implies that the profitability and overall performance of the bank
decreases with increasing GNPLR due to higher provisions for loan losses. Consequently, effective
credit risk management is indispensable for the long-term survival of the banking sector (Noor &
Siddiqui, 2019).

An unchecked GNPLR can drain a bank’s capital reserves, which affects lending ability: as these loans
dwarf income generation, it disenchants investors–all of which is antithetical to profitability. Banks
must remain ever more vigilant in risk management and stay ahead of these loans to pursue a
financially robust position that sustains profitability (Shahin et al., 2022). Finally, the relationship
between GNPLR and the bank’s performance is instructive on two main counts. Firstly, it is a helpful
guide to effective risk management and overall banking sector soundness. On this score, an assessment
of this association will aid in identifying areas of weakness and choosing how to address the
profitability effects of non-performing loans. Secondly, regulators and policymakers must implement
measures to enhance financial soundness and protect stakeholder interests (Matemane & Wentzel,
2019).

In conclusion, robust banking systems rely on effective risk management, of which GNPLR is a crucial
leading indicator. Empirical evidence demonstrates that elevated GNPLR is associated with more
extended periods of economic slowdown. However, banks with lower GNPLR report higher returns
and are less disrupted from their financial intermediation role. These are the very reasons why these
indicators matter to policy and regulation concerned with banking sector resilience and overall
economic health (Ibrahim, 2018).

Return on Assets (ROA) as a Measure of Bank Performance:

Return on assets (ROA) is a crucial metric reflecting a bank's profitability, i.e., its ability to generate
earnings from its assets. Data analyzed from Pakistan show a clear trend: banks that utilize credit risk
management strategies more effectively tend to have higher ROA values (Khan, 2014). ROA is more
than a metric; it is a story. A high ROA suggests that a bank is successfully earning on its assets. For
stakeholders reviewing a bank's financial performance, ROA provides significant intelligence on the
financial health of a bank and its proficiency in managing credit risks (Farah, 2020).

Achieving a high ROA is not enough. Banks must also remain dominant. Banks with dismal ROA
cannot grow complacent. They must enhance their asset quality, shrink non-performing loans, and
become more operationally nimble (Sharma, 2019).

222
Ali & Ahmed JEBV 4(1) 2024, 217-240

Diversification, too, is essential. New business ventures can raise profitability and revenue streams
(Alzoubi et al., 2022). For the banks of Pakistan, then, the lesson stands. Their future revolves around
innovation. The intelligent adoption of data analytics, regular credit risk model updates, and strict
adherence to industry best practices can change their future (Mushtaq et al., 2015). Furthermore,
insight into the value added from the ROA and ROI matrices and a thorough review of historical trends
could likely be invaluable. At the same time, robust internal controls alongside an ongoing and
rigorous audit regimen could help them raise the quality of their risk management practices and ensure
those (Purnomo et al., 2022).

Capital Adequacy Ratio (CAR) and Bank Stability:

Against this backdrop, the banking metrics adjudicators prefer the Capital Adequacy Ratio (CAR) to
measure the bank’s resilience against future adversities. In this regard, the Pakistan State Bank points
out that the banks would have remained unfazed by unscheduled insomnia had they met the correct
ratio (Rahman et al., 2018). The capital adequacy ratio is significantly related to bank stability in
Pakistan’s banking sector as a solid capital base temporizes the risks (Rahman et al., 2018).

Critics argue that a high CAR does not guarantee risk readiness, as demonstrated by Lehman's crisis
(Connerty, 2010). This underscores the need for a comprehensive risk assessment beyond CAR (Spina,
2013). To safeguard the global financial system, the analysis must extend beyond CAR to include
liquidity ratios, leverage ratios, and stress testing (Kibritcioglu, 2002). Qualitative aspects such as
governance and risk culture enhance the evaluation process, creating a comprehensive risk
management blueprint (Tongurai & Vithessonthi, 2020).

Macroeconomic Factors: Inflation, GDP, and Interest Rates:

Different macroeconomic factors significantly impact the credit risk landscape and the financial well-
being of institutions (V. De Leon, 2020; GRISSE, 2021). For instance, high inflation reduces the value
of loans, causing the risk of default to increase; GDP growth generally improves credit quality and
reduces default rates. Fluctuations in interest rates impact borrowing costs and the likelihood of default
(Caruso & Coroneo, 2023).

The State Bank of Pakistan's policies, such as adjustments in interest rates, influence collateralized
loaning and loan repaying ability (Platte, 2022). Securitization, if used prudently, could maximize
wealth, but default risks are to be minimized (Kashyap, 2016). In Pakistan, GDP growth fortifies
banking operations while inflation weakens them. The interest rate changes either encourage or
discourage borrowing; thus, the lending policies are based on this factor. Knowing these dynamics of
banks and borrowers helps the banks adjust to the changes in economic conditions and, therefore,
remain protected from defaults (Zubair & Sajid, 2014).

223
Ali & Ahmed JEBV 4(1) 2024, 217-240

RESEARCH METHODOLOGY

In this study, we scrutinize the nuanced undertaking of credit risk management and performance
of commercial banks in a developing country, i.e., Pakistan, in the presence of macroeconomic factors.
The methodological architecture of the study is based on panel Vector Autoregressive (VAR)
techniques. This approach provides a sagacious analytical framework for multivariate time series data.
This allows for simultaneous examination of the endogenous relationship among variables over time
when we have multiple banks for a critical dependent variable. The model specification is expressed
as:

𝒀𝒊𝒕 = 𝑨𝟎 + 𝑨𝟏 𝒀𝒊𝒕−𝟏 + ⋯ + 𝑨𝒑 𝒀𝒊𝒕−𝒑 + 𝜺𝒊𝒕

Within this model, 𝒀𝒊𝒕 represents a vector of endogenous variables, amalgamating bank-centric
indicators and macroeconomic metrics. The indices 𝒊 and 𝒕 demarcate individual banking entities and
specific time intervals.
To bolster our analysis, based on the foundational empirical model, we derived the following model:

𝒀𝒊𝒕 = 𝑨(𝑳)𝒀𝒊𝒕 + 𝜺𝒊𝒕 ,

Where,

𝒀𝒊𝒕 = [𝑩𝒂𝒏𝒌 𝑺𝒑𝒆𝒄𝒊𝒇𝒊𝒄 𝑽𝒂𝒓𝒊𝒂𝒃𝒍𝒆𝒔, 𝑴𝒂𝒄𝒓𝒐𝒇𝒊𝒏𝒂𝒏𝒄𝒊𝒂𝒍 𝑽𝒂𝒓𝒊𝒂𝒃𝒍𝒆𝒔]

𝒀𝒊𝒕 = [𝑮𝑵𝑷𝑳𝑹𝒊𝒕 , 𝑹𝑶𝑨𝒊𝒕 , 𝑪𝑨𝑹𝒊𝒕 , 𝒚𝒊𝒕 , 𝝅𝒊𝒕 , 𝒓𝒊𝒕 , 𝑫𝒖𝒎𝒎𝒊𝒆𝒔𝒊𝒕 ],

Where macroeconomic variables𝑦𝑖𝑡 , 𝑎𝑛𝑑 𝑟𝑖𝑡, respectively, are the real growth, Consumer Price
inflation, and 3-month Treasury bill rate. Our empirical focus is pivoted around a meticulously curated
set of variables, informed by pertinent literature and with direct bearing on the study's objectives.
These variables spanned bank profitability metrics, liquidity indicators, capital adequacy, etc. A cohort
of 18 strategically significant commercial banks in Pakistan was selected for examination, with
choices informed by market footprint, their emblematic nature within the broader banking milieu, and
the reliability of the data they proffered.

In consonance with the open source philosophy, the data frame set for our exploration spans a quarterly
time series between 2010Q1 and 2022Q4. The financial statements, the principal manuscript for this
inquisition, were methodically distilled from eminently authentic repositories, including official bank
domains, annual discourses, and pertinent regulatory sanctums. Despite employing insider techniques,
the process was not without its tribulations. Discontinuities, for instance, necessitated contingencies,
such as direct outreach to banks. Also, as with legacy, the reporting standards varied, mandating hours
of normalization— a chore augmented by the mutable accounting standards and regulatory mandates.
Robust validation infrastructure, however, was instituted so as not once to forsake exactness or
reproducibility. The objective of this latest illumination is a holistic view of the inchoate theme, as
viewed through the refined lens of our VAR framework, aimed to juxtapose the credit risk
224
Ali & Ahmed JEBV 4(1) 2024, 217-240

management imperative versus the macroeconomic health of Pakistan's commercial banks. Based on
this model, we anticipate significant scholarly and practical contributions from our research.

RESULT AND DISCUSSION

Data Analysis and Interpretation

To investigate the presence of unit roots within the panels, we conducted the Im-Pesaran-Shin
(IPS) test for all the variables. The IPS test employs autoregressive parameters and fixed-N critical
values for panel-specific. The hypotheses for IPS tests are:

H₀: All panels have unit roots

Hₐ: Some panels have stationarity

The results of the IPS tests for the variables are as follows:
Variable Statistic p-value Critical Values (1%, 5%, 10%) Result

GNPLR 𝑍𝑡̃𝑏𝑎𝑟 : 1.6326 0.9487 -1.980, -1.840, -1.780 Likely non-stationary

ROA 𝑍𝑡̃𝑏𝑎𝑟 : -6.0647 6.611e-10 -1.980, -1.840, -1.780 Likely stationary
CAR : -1.0317 0.1511 - Marginally non-stationary
RGDPG 𝑍𝑡̃𝑏𝑎𝑟 : -6.7620 6.803e-12 -1.980, -1.840, -1.780 Likely stationary
INFL 𝑍𝑡̃𝑏𝑎𝑟 : 10.8341 1 -1.980, -1.840, -1.780 Likely non-stationary
TB3 𝑍𝑡̃𝑏𝑎𝑟 : 4.4093 0.99999482 -1.980, -1.840, -1.780 Likely non-stationary

These results give a sense of the time series characteristics in the variables and will be used as a basis
for the subsequent modeling and analysis. The lags used in these tests were selected automatically
based on the lowest AIC. Time series properties of the time series are of no concern as we are using
this data merely to establish that the variations are significantly different.
Non-stationary Variables: Transforming for Stationarity
Variable stationarity is necessary for econometric studies to draw reliable statistical inferences and
modeling outcomes. A variable is defined as non-stationary if it displays trends. Also, non-stationary
variables are called unit root variables. The existence of unit roots or trends can result in the creation
of spurious relationships, and it can complicate analysis. To expunge trends and make the variables
stationary for further analysis, the Im-Pesaran-Shin test was used to confirm the presence of non-
stationary variables (inflation, interest rate, CAR, GNPLR) and differentiate these into stationary
series by differencing. A subsequent first differencing resolves the presence of unit roots.
Implementing the sequence ensures that the dataset is ready for a vehement test and affords accurate
statistical inferences in the later stages.

225
Ali & Ahmed JEBV 4(1) 2024, 217-240

Cointegration Analysis: Insights from the Kao (1999) Test

Cointegration analysis is an instrumental tool in econometric studies, serving as a lens to discern long-
term associations between variables. Such an analysis probes whether a set of variables exhibit
synchronous movement over extended periods (long run), potentially reflecting shared intrinsic
relationships. Within the framework of this study, the Kao test was employed to evaluate the
Cointegration stance of our panel variables.
Applying the Kao test, the competing hypotheses set the stage: the null hypothesis (Ho) contending
the absence of cointegration, and the alternative hypothesis (Ha) proposing cointegration across all
panels. This evaluative exercise spanned a panel dataset composed of 18 panels, boasting an average
period count of 48.778.
The essence of the panel variables' cointegration profile is encapsulated in the outcomes of the Kao
test, summarized in Table I.
Table I: Kao Test Findings on Cointegration
Kao Test for Cointegration
Ho: No cointegration Number of panels = 18
Ha: All panels are cointegrated. Avg. number of periods = 48.778

Cointegrating vector: Same

Panel means: Included Kernel: Barlett
Time trend: Not included Lags: 2.28 (Newey-West)
AR parameter: Same Augmented lags: 1

Statistic p-value
Modified Dickey-Fuller t -0.3169 0.3757
Dicky-Fuller t 0.2171 0.4140
Augmented Dicky-Fuller t 3.3545 0.0004
Unadjusted modified Dicky-Fuller t -1.7795 0.0376
Unadjusted Dicky-Fuller t -0.7186 0.2362

Insights gained from the test statistics and associated p-values provide valuable information on the
relationship dynamics among panel variables. The Modified Dickey-Fuller and Dickey-Fuller t
statistics indicate a weak indication of cointegration. In contrast, the Augmented Dickey-Fuller t
statistic presents strong evidence supported by a low p-value.

In contrast, the Unadjusted Modified Dickey-Fuller t statistics provide more cautious support for
cointegration. The Kao test results also offer conflicting evidence on the cointegration question,
suggesting that further research and nuanced interpretation of these results is required. One direction
of such future research might involve a probe of the underlying long-term connections and their
implications for constructing econometric models and predicting financial markets.

226
Ali & Ahmed JEBV 4(1) 2024, 217-240

Dissecting Dynamics of Stationary Variables

Because the interactions of economic variables are essential for policy and research, this paper uses a
framework to study such interactions. Our framework combines the Generalized Method of Moments
(GMM) estimation with Panel Vector Autoregression (PVAR). We employ 18 panels from 860
observations that are quarterly for each panel (i.e., 18*860= 15480 observations). Our implementation
allows an extensive examination of variable dynamics and, as evident in the GMM criterion, results
in a practically precisely identified model. A model is precisely identified by three moments per
equation; an under-identified model has fewer moments than equations. Our robust GMM weight
matrix is based on the homoskedasticity and serial correlation of the errors. The lags of the endogenous
variables serve as instruments to address endogeneity.

These analyses contribute to the panel data analytics literature. Moreover, they underscore the need to
understand economic coordination. Our methodology depends on GMM weight matrix robustness and
the use of instrumental variables. Appendix I reports detailed results such as estimated coefficients,
their standard errors, the t-statistics, the value of the GMM criterion, and the lags of the endogenous
variables.
Model Stability

It is essential to ensure the meaningfulness and accuracy of the results yielded by a panel VAR model.
Assessing the stability of the model has led us to verify all of its eigenvalues falling within the unit
circle, confirming its stability with the Eigenvalue stability criterion. Eigenvalues are scaling factors
used in a vector transformation. Stability is guaranteed when its values are less than or equal to one.
Our investigation revealed that all eigenvalues have moduli less than one. This signified their stability.
The maximum eigenvalue, while closely approaching one, remains within the bounds of stability. The
fact that all eigenvalues remain within the unit circle confirmed the model's stability. It is a self-
correcting system, immune to wild trends. Its ability to be ratified reassures us of its reliability.
Table II: Eigenvalue Stability Condition
Eigenvalue
Modulus
Real Imaginary
0.998 0 0.998
0.666 0.418 0.786
0.666 -0.418 0.786
-0.372 -0.682 0.777
-0.372 0.682 0.777
0.716 0 0.716
-0.278 0.13 0.307
-0.278 -0.13 0.307
0.281 0 0.281
-0.259 0 0.259
0.138 -0.037 0.142
0.138 0.037 0.142

227
Ali & Ahmed JEBV 4(1) 2024, 217-240

Relationship between Variables: Granger Causality Tests

We turn to results from our benchmark panel VAR model. First, we present the results from the
Granger-causality test in Table III.
Table III: Granger Causality Test
Equation
GNPLR ROA CAR RGDP INFL TB3
Excluded
GNPLR 0.098 0.318 0.047 0.044 0.978
ROA 0.246 0.144 0.415 0.468 0.975
CAR 0.099 0.033 0.881 0.663 0.269
RGDP 0.297 0.367 0.734 0.000 0.037
INFL 0.416 0.354 0.870 0.002 0.000
TB3 0.006 0.496 0.086 0.299 0.000
ALL 0.003 0.005 0.045 0.000 0.000 0.000
Notes: (i) H0: Excluded variable Does No Granger Cause Equation Variable. (ii) p-values are reported. (iii) Bold entries
show rejection at a 10 percent confidence level.

Examining the Relationship of Credit Risk with Other Variables:

The study focused on the relationships between credit risk and various economic determinants.
Specifically, variables such as credit risk (GNPLR), profitability (ROA), capital adequacy ratio
(CAR), real growth rate (𝑦), inflation (𝜋), and interest rates (𝑟) were investigated. Table 3 shows that
for credit risk, all variables are jointly significant. Capital adequacy and interest rates individually
Cause credit risk in the banking sector. This implies that changes in a bank's solvency and the prevalent
price of money in the market can meaningfully predict the credit riskiness of the banks.
Investigating Profitability's Interactions with Other Variables:
The banks' profitability is significantly predictable by the credit risk and solvency of the banks.
However, the macroeconomic variables are not significant. Nonetheless, all variables jointly
significantly predict profitability. A rising level of credit risk puts a drag on profitability by increasing
the provisioning expenses of the banks, affects the solvency, and leads to lower profits.
Relationship between Capital Adequacy Ratio and Other Variables:
Capital adequacy is only significantly predicted by the level of economic interest rates. While
individually insignificant, all macro and bank-specific variables are jointly significant.
Relationship between Real Growth Rate and Other Variables:
It is interesting to note that real economic growth is Granger-caused by the credit risk and the inflation
levels. It is well known that inflation causes uncertainty in the economy and affects aggregate demand
by causing households and firms to postpone consumption and investment decisions. At the same time,
a rising credit risk in the banks would restrain banks from making loans to households and firms, thus
hampering the overall investment levels in the economy.

228
Ali & Ahmed JEBV 4(1) 2024, 217-240

Relationship between Inflation and Other Variables:

The changes in price levels are affected by the changes in economic activity, the interest rates as well
as the credit risk of banks. The channels seem to be economically consistent and as per expectations.
Theoretically, the price levels rise or fall when the economy operates above or below its potential.
Further, a rising cost of investments (interest rates) would hamper productivity and inflation.
Furthermore, when banks brake on lending activity due to a rise in credit riskiness, the investment and
economic activity affect the price levels.
Relationship between Interest Rates and Other Variables:
The interest rates, as expected, are driven by macroeconomic factors, viz., real growth and inflation
levels. The relationship can be explained via essential interaction within the IS-LM framework and
the well-known Fisher effect.
Impulse Response Analysis of Panel VAR Model: Dynamic Interactions among Variables
Having established the relationship among various variables via the Granger Causality results, we
delve into the results derived from the impulse response analysis conducted on our benchmark panel
VAR model. Impulse response functions (IRFs) enable a nuanced understanding of the dynamic
interplay and implications of variations in one variable on others within the model. This investigation
is central to deciphering the transmission mechanisms and interconnectedness inherent in the panel
VAR structure. The focal variables for this analysis are the credit risk (GNPLR) about shocks in the
macro-variables like interest rate (TB3), real output growth (RGDPG), and inflation changes (DINFL),
as well as bank-specific determinants, such as profitability (ROA) and alterations in solvency (DCAR).
Response of Credit Risk to Monetary Policy Shock:

Utilizing the IRFs for the changes in interest rate (𝑟𝑖𝑡 ), we gauge the implications of fluctuations in
this variable on the credit risk (GNPLR). A graphical representation in Figure 1 elucidates the
dynamics over 15 quarters. Notably, an upswing in interest rates amplifies credit risk. To elaborate, a
contractionary monetary policy pushes the cost of borrowing up for new and existing loans, increasing
the propensity of loan impairments, specifically non-performing loans. This escalation unfolds
progressively, peaking at approximately eight quarters, in line with the conventional wisdom that the
repercussions of a tight monetary policy manifest after a delay.

229
Ali & Ahmed JEBV 4(1) 2024, 217-240

Figure 1: Response of Credit Risk to a Monetary Policy Shock

Response of Credit Risk to Economic Growth:

A careful analysis of the graph in Figure 2 reveals the complex relationship between output growth
and loan impairments. An increase in economic activity has an inverse effect on the non-performance
of loans. This inverse relationship is based on the idea that a thriving economy leads to higher income
levels, making it easier for borrowers to service and repay loans, thus reducing non-performing loans.
Another perspective suggests that the decrease in the non-performing loan ratio (GNPLR) is due to a
consistent level of non-performing loans and an increased lending velocity, which could mitigate the
contagion levels in loans.

Figure 2: Response of Credit Risk to a Shock in Economic Activity

230
Ali & Ahmed JEBV 4(1) 2024, 217-240

Response of Credit Risk to Inflation:

Our analysis further examines the relationship between credit risk (GNPLR) and inflationary
pressures. Figure 3 shows that changes in inflation have an impact on credit risk. An increase in
inflation generally leads to higher credit risk. This finding highlights the potential adverse effect of
inflation on borrowers' ability to repay, resulting in a rise in non-performing loans and an overall
increase in credit risk. The increase in credit risk due to inflation can be attributed to various factors.
Inflation primarily reduces economic agents' purchasing power, limiting their ability to service loans.
Additionally, inflation introduces economic uncertainty, making cash flow planning more complex
for borrowers. This increased uncertainty potentially raises credit risk. It is important to note that the
effects of inflation on credit risk may not be immediate but rather unfold gradually, as depicted in the
graph. However, these impacts diminish after the completion of the seven quarters.

Figure 3: Response of Credit Risk to a Shock in Inflation

Response of Credit Risk to Profitability:

Verifying the impact of credit risk on profitability, or vice versa, is also essential. As revealed in
Figure 4, an increase in profitability leads to a significant decrease in bank credit risk. In contrast, a
reduction in bank profitability increases credit risk. Diminishing bank profitability can signal trouble
for banks in generating revenue to cover costs and potential loan losses. Causes of a reduction in bank
profitability may include higher credit impairment provisions, increased loan defaults, and challenging
economic conditions. Hence, worsening credit risk looms, reflecting prospective degradation of loan
quality and increased likelihood of non-performing loans. Importantly, this relationship is bi-
directional; as declining profitability potentially leads to credit risk, elevated credit risk can decrease
bank profitability.

231
Ali & Ahmed JEBV 4(1) 2024, 217-240

Figure 4: Response of Credit Risk to a Shock in Profitability

Response of Credit Risk to Capital Adequacy Ratio (CAR):

Figure 5 offers a more detailed correlation analysis of credit risk (GNPLR) and the Capital Adequacy
Ratio (CAR). The figure shows that an increase in the CAR resulted in a decline in credit risk. The
CAR measures a bank’s capital strength and ability to absorb losses. A decline in the CAR signifies
that a bank has weaker capital reserves, rendering its loan portfolio more risky. This decline in the
CAR, associated with an increase in credit risk, signifies a heightened probability of loan defaults and
impairments. This observation validates the widely accepted argument that a robust capital position
fortifies a bank to endure shocks.
Conversely, a bank with lower capital levels is more exposed to loan defaults and a deterioration in
credit quality. Moreover, the Granger causality results confirm that the relationship between credit
risk and CAR is unidirectional: a fall in CAR raises credit risk, whereas an increase in credit risk will
strain bank capital metrics, leading to higher loan impairment provisions and potential capital
depletion.

Figure 5: Response of Credit Risk to a Shock in Capital Adequacy

232
Ali & Ahmed JEBV 4(1) 2024, 217-240

CONCLUSION AND POLICY IMPLEMENTATION

Robust credit risk management strategies are vital to the resilience and performance of the
banking sector, and this is especially the case in emerging economies like Pakistan. This paper
provides a comprehensive evolutionary examination of the intricate relationships between credit risk
management and the operational dynamics of commercial banks in the context of Pakistan’s
dynamically evolving financial landscape. We deploy panel Vector Autoregressive (VAR)
methodology to explore the complex relationships among many variables represented in multivariate
time series data, and we thus focus on the endogenous relationships between bank-specific indicators
and the broader macroeconomic determinants.
Our analytical approach, substantiated by the empirical model, incorporated macro financial and micro
(bank-specific) variables, making our approach comprehensive. The diversity and number of our
variables (looking at profitability, liquidity, etc.) demonstrate the depth of our study and the fact that
we have also aligned our study to the current discourse. We have carefully examined a select group of
18 central commercial banks so that our findings truly reflect Pakistan's banking sector, capturing its
nuances, challenges, and potential.
This research culminates a comprehensive picture that underscores the cardinal significance of credit
risk management in the health and resilience of banks and the interconnectedness of micro-level
banking operations with broader macroeconomic undulations, especially in a dynamic economic
environment such as Pakistan.
Our results indicate that credit risk is influenced not only by bank-specific variables, including
profitability and solvency, but also significantly by monetary policy, economic activity, and inflation.
However, while the effect of shocks from micro-level idiosyncratic variables on credit risk is
instantaneous, macro-level systemic variables are observed with a lag, as the extant literature suggests.
This study translates these insights to create value, transcending an academic exercise. It guides
policymakers, financial practitioners, and banking leaders to inform their credit risk strategies amid
economic ebb and flow. It suggests a roadmap for future academic work, combining rigorous research
with practical relevance to indicate that there is more than a token to the argument that scholarly work
can influence - and improve - real-world financial landscapes.

REFERENCES
Alazis, M. (2020). EFFECT OF CAR, LDR, ROA, ROA, AND NIM TOWARD THE
COMMERCIAL BANK IN INDONESIA. International Journal of Economics, Business and
Accounting Research (IJEBAR), 4(01). https://doi.org/10.29040/ijebar.v4i01.954
Alzoubi, M., Alkhatib, A., Abdalmajeed Alsmadi, A., & Kasasbeh, H. (2022). Bank size and capital:
A trade-off between risk-taking incentives and diversification. Banks and Bank Systems, 17(4), 1–
11. https://doi.org/10.21511/bbs.17(4).2022.01

233
Ali & Ahmed JEBV 4(1) 2024, 217-240

Ando, S. (2019). International Financial Connection and Stock Return Co-movement. IMF Working
Papers, 16(141). https://doi.org/10.5089/9781513509822.001
Arshed, N., Sohail, H., & Gulzar, M. (2023). Investigating the Institutional Quality Integration with
Islamic Banking Development in Promoting Entrepreneurship. Journal of Entrepreneurship and
Business Venturing, 3(1). https://doi.org/10.56536/jebv.v3i1.25
Arora, R. (2021). Driving excellence in credit risk management practices in commercial lending - an
empirical analysis of Indian public sector banks. International Journal of Business Continuity and
Risk Management, 11(1), 1. https://doi.org/10.1504/ijbcrm.2021.10036616
Barjaktarović, L., Vesić, T., & Laki, B. (2022). What can be expected in credit-risk management from
NPL in the western Balkans region in the future? International Review, (3–4), 96–103.
https://doi.org/10.5937/intrev2204100b

Bhatt, T. K., Ahmed, N., Iqbal, M. B., & Ullah, M. (2023). Examining the determinants of credit risk
management and their relationship with the performance of commercial banks in Nepal. Journal
of Risk and Financial Management, 16(4), 235. https://doi.org/10.3390/jrfm16040235
CARUSO, A., & CORONEO, L. (2023). Does Real‐Time Macroeconomic Information Help to
Predict Interest Rates? Journal of Money, Credit and Banking.
https://doi.org/10.1111/jmcb.13021
Connerty, A. (2010). The credit crunch: the collapse of Lehman Brothers - and a Hong Kong scheme
to handle Lehman claims. Amicus Curiae, 7–12. https://doi.org/10.14296/ac.v2010i83.1224
Erdoğdu. (2017). Assessing the Impact of Non-Performing Loans on Economic Growth in Turkey.
American Research Journal of Business Management. https://doi.org/10.21694/2379-1047.16007
Farah, A. (2020). Determinants of Bank Performance: ROE, ROA Profitability Measures Amongst
Islamic and Conventional Banks in the GCC. SSRN Electronic Journal.
https://doi.org/10.2139/ssrn.3579883
GRISSE, C. (2021). Lower Bound Uncertainty and Long‐Term Interest Rates. Journal of Money,
Credit and Banking, 55(2–3), 619–634. https://doi.org/10.1111/jmcb.12890
Gunningham, N. (2020). A Quiet Revolution: Central Banks, Financial Regulators, and Climate
Finance. Sustainability, 12(22), 9596. https://doi.org/10.3390/su12229596
Han, C., & Kang, J. (2008). An extended CreditRisk+ framework for portfolio credit risk management.
The Journal of Credit Risk, 4(4), 63–80. https://doi.org/10.21314/jcr.2008.080

Harb, E., El Khoury, R., Mansour, N., & Daou, R. (2022). Risk management and bank performance:
Evidence from the MENA region. Journal of Financial Reporting and Accounting, 21(5), 974–
998. https://doi.org/10.1108/jfra-07-2021-0189
Ibrahim, I. (2018). The Relationship Between Liquidity Risk and Bank Performance in Bank: Oversea-
Chinese Banking Corporation Limited, Bank in Singapore. SSRN Electronic Journal.
https://doi.org/10.2139/ssrn.3302063
Kashyap, R. (2016). Securities Lending Strategies, Valuation of Term Loans using Option Theory.
SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2826485

234
Ali & Ahmed JEBV 4(1) 2024, 217-240

Khan, Q. (2014). Credit Risk Management in Islamic and Conventional Banks (An Empirical
Investigation from Pakistan). SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3310956
Khurram, M., & Siddiqui, D. A. (2021). The Impact of Control Quality on the Non-Performing Loans
of Pakistan Listed Banks. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3942284
Kibritcioglu, A. (2002). Excessive Risk-Taking, Banking Sector Fragility, and Banking Crises. SSRN
Electronic Journal. https://doi.org/10.2139/ssrn.317491
Lafuente, E., Vaillant, Y., & Vendrell-Herrero, F. (2019). Conformance and performance roles of bank
boards: The connection between non-performing loans and non-performing directorships.
European Management Journal, 37(5), 664–673. https://doi.org/10.1016/j.emj.2019.04.005
Lavrushin, O., & Sokolinskaya, N. (2020). Retraction: Confidence level and credit risk analysis in
Russian banks. Banks and Bank Systems, 15(2), 38–46.
https://doi.org/10.21511/bbs.15(2).2020.04
Matemane, R., & Wentzel, R. (2019). Integrated reporting and financial performance of South African
listed banks. Banks and Bank Systems, 14(2), 128–139.
https://doi.org/10.21511/bbs.14(2).2019.11
Mennawi, A. N. A. (2020). The Impact of Liquidity, Credit, and Financial Leverage Risks on Financial
Performance of Islamic Banks: A Case of Sudanese Banking Sector. Risk and Financial
Management, 2(2), p59. https://doi.org/10.30560/rfm.v2n2p59
Mushtaq, M., Ismail, A., & Hanif, R. (2015). Credit Risk, Capital Adequacy and Banks Performance:
An Empirical Evidence from Pakistan. International Journal of Financial Management, 5(1).
https://doi.org/10.21863/ijfm/2015.5.1.001
Natufe, O. K., & Evbayiro-Osagie, E. I. (2023). Credit Risk Management and the Financial
Performance of Deposit Money Banks: Some New Evidence. Journal of Risk and Financial
Management, 16(7), 302. https://doi.org/10.3390/jrfm16070302
Noor, I., & Siddiqui, D. A. (2019). Evidence of Non-Linear Relationship between Non-Interest
Income and Profitability of Commercial Banks in Pakistan. Asian Journal of Economic Modelling,
7(1), 14–26. https://doi.org/10.18488/journal.8.2019.71.14.26
Nuta, A. C., Habib, A. M., Neslihanoglu, S., Dalwai, T., & Rangu, C. M. (2024). Analyzing the market
performance of Romanian firms: do the COVID-19 crisis and classification type
matter? International Journal of Emerging Markets. https://doi.org/10.1108/ijoem-05-2023-0842
Platte, D. (2022). Banks’ Management of Interest Rate Risk: Empirical Evidence from Savings Banks
and Credit Cooperatives. Zeitschrift Für Bankrecht Und Bankwirtschaft, 34(4), 228–239.
https://doi.org/10.15375/zbb-2022-0405
Ponomarenko, V., Kolodiziev, O., & Chmutova, I. (2017). Benchmarking of bank performance using
the life cycle concept and the DEA approach. Banks and Bank Systems, 12(3), 74–86.
https://doi.org/10.21511/bbs.12(3).2017.06
Purnomo, H., Nurcahyani, I. S., & Maryati, M. (2022). Does Car Influence RoA? Evidence from State-
Owned Banks. Jurnal Ilmu Keuangan Dan Perbankan (JIKA), 12(1), 34–41.
https://doi.org/10.34010/jika.v12i1.8232

235
Ali & Ahmed JEBV 4(1) 2024, 217-240

Rakhaev, V. A. (2020). Developing Credit Risk Assessment Methods to Make loss Provisions for
Potential loans. Finance: Theory and Practice, 24(6), 82–91. https://doi.org/10.26794/2587-5671-
2020-24-6-82-91
Rashid, A., Ahmad, F., & Yasmin, A. (2017). Exploring the relationship between macroeconomic
indicators and sovereign credit default swap in Pakistan. The Journal of Risk Finance, 18(4), 368–
380. https://doi.org/10.1108/jrf-03-2017-0049
Rehman, Z. U., Muhammad, N., Sarwar, B., & Raz, M. A. (2019). Impact of risk management
strategies on the credit risk faced by commercial banks of Balochistan. Financial Innovation, 5(1).
https://doi.org/10.1186/s40854-019-0159-8
Safar Nasir, M., Oktaviani, Y., & Andriyani, N. (2022). Determinants of Non-Performing Loans and
Non-Performing Financing Level: Evidence in Indonesia 2008-2021. Banks and Bank Systems,
17(4), 116–128. https://doi.org/10.21511/bbs.17(4).2022.10
Shahin, R., Khalil, M., & Sallam, H. (2022). The nexus between interest rate and bank profitability:
Does bank prudential capital matter? Banks and Bank Systems, 17(2), 113–123.
https://doi.org/10.21511/bbs.17(2).2022.10
Sharma, M. (2019). A Comparative Study on Return on Assets (ROA) and Return on Equity (ROE)
of The Three Biggest Banks of India. Journal of Development Research, 12(4), 13.
https://doi.org/10.54366/jdr.12.4.2019.13-19
Shrestha, R. (2017). The Impact of Credit Risk Management on Profitability: Evidence from Nepalese
Commercial Banks. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2938546
Spina, J. (2013). The Failure of Lehman Brothers. SSRN Electronic Journal.
https://doi.org/10.2139/ssrn.2337124
Spuchlakova, E., & Misankova, M. (2017). Risk management of Credit Default Swap. New Trends
and Issues Proceedings on Humanities and Social Sciences, 3(4), 229–234.
https://doi.org/10.18844/gjhss.v3i4.1573
Tongurai, J., & Vithessonthi, C. (2020). Bank regulations, bank competition, and bank risk-taking:
Evidence from Japan. Journal of Multinational Financial Management, 56, 100638.
https://doi.org/10.1016/j.mulfin.2020.100638
Ur Rahman, K., Akram, H., & Noreen, A. (2018). Concentration Risk Analysis of Commercial Banks
and its Relation with GDP and GNI: Comparison of Islamic and Conventional Banks in Pakistan.
COMSATS Journal of Islamic Finance, 3(2), 88–103. https://doi.org/10.26652/cjif.3201825
V. De Leon, M. (2020). The impact of credit risk and macroeconomic factors on profitability: the case
of the ASEAN banks. Banks and Bank Systems, 15(1), 21–29.
https://doi.org/10.21511/bbs.15(1).2020.03
Zubair, M., & Sajid, M. (2014). Impact of Credit Risk Management on Bank Performance: An
Empirical Study from Commercial Banks of Pakistan Listed at Pakistan Stock Exchange (PSE).
SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3008939

236
Ali & Ahmed JEBV 4(1) 2024, 217-240

Appendix: I

Panel Vector Autoregressive Model Results

Coef. Std.Err. Z P>z [95%Conf. Interval]

gnplr
gnplr
L1. 0.734 0.133 5.54 0 0.474 0.994
L2. 0.259 0.09 2.88 0.004 0.083 0.434

roa
L1. -0.367 0.325 -1.13 0.259 -1.005 0.271
L2. 0.246 0.237 1.04 0.299 -0.218 0.71

dcar
L1. -19.524 12.286 -1.59 0.112 -43.603 4.556
L2. -12.938 6.017 -2.15 0.032 -24.731 -1.146

rgdpg
L1. -0.059 0.047 -1.25 0.213 -0.152 0.034
L2. -0.072 0.053 -1.36 0.175 -0.176 0.032

dinfl
L1. -0.054 0.095 -0.57 0.566 -0.24 0.131
L2. -0.097 0.082 -1.19 0.235 -0.257 0.063

dtb3
L1. -0.039 0.122 -0.32 0.751 -0.277 0.2
L2. 0.23 0.074 3.09 0.002 0.084 0.376
roa
gnplr
L1. -0.024 0.025 -0.95 0.344 -0.072 0.025
L2. 0.027 0.013 2.08 0.038 0.002 0.052

roa
L1. 0.857 0.168 5.1 0 0.528 1.187
L2. -0.021 0.095 -0.23 0.821 -0.207 0.164

dcar
L1. 3.275 2.553 1.28 0.2 -1.729 8.28
L2. 4.015 1.719 2.34 0.019 0.646 7.384

rgdpg
237
Ali & Ahmed JEBV 4(1) 2024, 217-240

L1. 0.002 0.014 0.13 0.897 -0.026 0.029

L2. 0.018 0.013 1.4 0.161 -0.007 0.043

dinfl
L1. 0.002 0.02 0.12 0.907 -0.038 0.043
L2. 0.023 0.021 1.1 0.269 -0.018 0.063

dtb3
L1. -0.006 0.028 -0.22 0.828 -0.061 0.049
L2. -0.024 0.021 -1.11 0.268 -0.066 0.018
dcar
gnplr
L1. 0.001 0.001 0.78 0.433 -0.001 0.003
L2. 0 0.001 0.52 0.604 -0.001 0.002

roa
L1. -0.01 0.005 -1.87 0.062 -0.02 0
L2. -0.002 0.003 -0.65 0.518 -0.009 0.004

dcar
L1. -0.332 0.12 -2.76 0.006 -0.567 -0.096
L2. -0.049 0.083 -0.59 0.555 -0.212 0.114

rgdpg
L1. -0.001 0.001 -0.73 0.463 -0.002 0.001
L2. 0 0.001 -0.03 0.973 -0.002 0.002

dinfl
L1. -0.001 0.001 -0.39 0.695 -0.003 0.002
L2. 0 0.001 -0.06 0.952 -0.002 0.002

dtb3
L1. -0.001 0.002 -0.6 0.549 -0.005 0.002
L2. 0.002 0.001 1.87 0.062 0 0.004
rgdpg
gnplr
L1. -0.396 0.211 -1.87 0.061 -0.81 0.018
L2. -0.034 0.132 -0.26 0.797 -0.293 0.225

roa
L1. 0.897 0.68 1.32 0.187 -0.437 2.231
L2. 0.078 0.473 0.16 0.87 -0.85 1.005

238
Ali & Ahmed JEBV 4(1) 2024, 217-240

dcar
L1. 5.743 12.946 0.44 0.657 -19.63 31.116
L2. 0.391 12.658 0.03 0.975 -24.419 25.2

rgdpg
L1. 0.55 0.107 5.16 0 0.341 0.758
L2. -0.416 0.129 -3.24 0.001 -0.668 -0.165

dinfl
L1. -0.72 0.205 -3.5 0 -1.122 -0.317
L2. -0.486 0.168 -2.89 0.004 -0.816 -0.156

dtb3
L1. 0.387 0.264 1.47 0.142 -0.13 0.905
L2. 0.109 0.166 0.66 0.512 -0.216 0.434
dinfl
gnplr
L1. -0.256 0.152 -1.68 0.093 -0.555 0.043
L2. -0.075 0.095 -0.79 0.427 -0.261 0.111

roa
L1. 0.658 0.544 1.21 0.227 -0.409 1.725
L2. 0.074 0.388 0.19 0.848 -0.686 0.834

dcar
L1. 9.598 10.793 0.89 0.374 -11.555 30.752
L2. 6.004 9.963 0.6 0.547 -13.524 25.531

rgdpg
L1. -0.499 0.077 -6.45 0 -0.651 -0.347
L2. 0.146 0.105 1.38 0.166 -0.061 0.352

dinfl
L1. -0.17 0.163 -1.04 0.297 -0.49 0.15
L2. -0.503 0.131 -3.83 0 -0.76 -0.245

dtb3
L1. 1.219 0.193 6.33 0 0.842 1.597
L2. 0.291 0.128 2.28 0.023 0.04 0.542
dtb3
gnplr

239
Ali & Ahmed JEBV 4(1) 2024, 217-240

L1. -0.008 0.039 -0.2 0.842 -0.084 0.068

L2. -0.001 0.012 -0.11 0.909 -0.024 0.022

roa
L1. 0.021 0.168 0.12 0.903 -0.309 0.35
L2. -0.015 0.093 -0.16 0.873 -0.196 0.167

dcar
L1. -4.427 2.792 -1.59 0.113 -9.899 1.045
L2. -1.563 2.148 -0.73 0.467 -5.773 2.647

rgdpg
L1. 0.082 0.046 1.76 0.078 -0.009 0.172
L2. 0.064 0.036 1.75 0.079 -0.007 0.135

dinfl
L1. 0.155 0.048 3.27 0.001 0.062 0.248
L2. -0.059 0.057 -1.05 0.295 -0.171 0.052

dtb3
L1. 0.402 0.077 5.19 0 0.25 0.554
L2. 0.064 0.047 1.34 0.179 -0.029 0.157

240