Economic Policy Uncertainty (EPU) has emerged as a critical factor influencing the financial landscape, particularly in the banking sector. The interplay between EPU, credit risk, and lending decisions is essential for understanding how banks navigate economic fluctuations and regulatory changes. This paper examines the impact of EPU on credit risk and lending decisions of banks, focusing on those listed on the Tehran Stock Exchange.

The significance of EPU in shaping banking behavior has been well-documented in various studies. For instance, research indicates that heightened economic uncertainty prompts banks to adopt more conservative lending practices, thereby affecting the overall credit supply within the economy [1]. This conservative behavior can lead to an increase in credit risk as banks become more selective in their lending criteria, which may ultimately hinder economic growth by limiting access to finance for businesses and individuals [2]. In the context of Iran, where economic volatility and policy shifts are prevalent, understanding the dynamics between EPU and banking operations is particularly crucial. Previous studies have shown that banks operating in environments characterized by high uncertainty tend to experience fluctuations in their performance metrics, including loan size and credit risk profiles [3]. Furthermore, the response of banks to EPU may vary based on their ownership structure, with state-owned banks potentially exhibiting different vulnerabilities compared to private banks [4].

This study utilizes a comprehensive dataset from 12 banks listed on the Tehran Stock Exchange over a defined period from 2019 to 2023. By employing the Generalized Method of Moments (GMMs) for data analysis, the research aims to elucidate the relationship between EPU, credit risk, and lending decisions. The findings are anticipated to contribute valuable insights into how Iranian banks can better manage risks associated with EPU while making informed lending decisions.

In conclusion, this paper seeks to fill a gap in the existing literature by providing empirical evidence on the effects of EPU on credit risk and lending behaviors within the Iranian banking sector. The implications of this research extend beyond academic discourse, offering practical recommendations for policymakers and banking institutions aiming to enhance financial stability amidst uncertain economic conditions.

Theoretical foundations play a crucial role in understanding the relationship between EPU, credit risk, and lending decisions of banks. This paper aims to explore these interconnections, drawing upon established theories and empirical evidence. EPU refers to the unpredictability surrounding government policies that can affect economic performance, including fiscal, monetary, and regulatory changes [1]. The credit channel theory serves as a foundational framework for this study, positing that fluctuations in economic conditions influence banks' lending behaviors and risk assessments [5]. According to this theory, when EPU rises, banks may become more conservative in their lending practices due to heightened perceived risks associated with uncertain economic environments [6].

The impact of EPU on credit risk can be understood through two primary mechanisms: direct and indirect. The direct mechanism involves how uncertainty affects banks' balance sheets. Increased EPU can create noise in expected income signals, complicating banks' ability to predict future economic conditions and leading to reduced loan supply [7]. Conversely, the indirect mechanism focuses on how EPU influences corporate behavior. When uncertainty rises, firms may increase cash reserves and delay investments due to anticipated risks, subsequently reducing their demand for bank credit [8], [9].

Empirical studies have consistently shown that higher levels of EPU correlate with decreased lending activity. For instance, Calmès and Théoret [10] found that macroeconomic uncertainty inhibits the scale of bank loans in the Canadian banking sector. Similarly, Talavera et al. [11] demonstrated a negative correlation between economic uncertainty and bank credit in Ukraine. These findings align with Bernanke's real options theory [12], which suggests that firms are likely to postpone investments in uncertain environments, thereby affecting their borrowing needs.

Moreover, the ownership structure of banks also plays a significant role in how EPU influences lending decisions. Research indicates that state-owned banks may respond differently to EPU compared to private banks. State-owned banks often possess greater financial stability and access to government support, allowing them to maintain lending levels even during periods of high uncertainty [13]. In contrast, private banks may adopt more cautious lending practices under similar conditions.

In conclusion, the theoretical framework for examining the impact of EPU on credit risk and lending decisions encompasses various theories and empirical insights. This study aims to contribute to this body of knowledge by providing an analysis specific to banks listed on the Tehran Stock Exchange.

The background of this research on the impact of EPU on credit risk and lending decisions of banks is grounded in a substantial body of literature that highlights the intricate relationships between these variables. Understanding these dynamics is crucial for both theoretical and practical implications in the banking sector. Orden-Cruz et al. [14] analyze the relationship between EPU and credit risk in U.S. commercial banks, revealing a statistically significant positive correlation. Their study emphasizes that less profitable and less solvent banks are particularly vulnerable to the impacts of EPU on credit risk.

A study by Ozili [15] investigates the correlation between EPU and Non Performing Loans (NPLs) across different regions, providing insights into how regional characteristics can explain variations in bank performance amidst uncertainty. The interplay between investment decisions and credit demand is further explored by Zabavnik and Verbič [16], who suggest that firms' investment hesitance due to uncertainty leads to reduced borrowing needs. Su et al. [17] explore how firms increase cash holdings during times of high EPU, which can further reduce their reliance on bank loans, thereby impacting the credit supply from banks. Research by Li and Hu [18] indicates that heightened EPU restrains overall credit growth through the bank lending channel, particularly affecting larger and riskier banks more severely than their more liquid counterparts.

In a study of Chinese commercial banks, Chi and Li [19] find that EPU negatively affects lending decisions, with a notable increase in NPLs during periods of high uncertainty. This highlights how regional variations can influence banking behavior under uncertainty. Baker et al. [1] define EPU as the unpredictability surrounding government policies that can significantly influence economic performance. Their findings suggest that increased EPU leads to conservative lending practices among banks, as they become more risk-averse in uncertain environments. Wang and Duan [20] construct an economic uncertainty index using GARCH models to demonstrate that economic uncertainty negatively correlates with credit growth, reinforcing the idea that banks must adapt to maintain stability. Calmès and Théoret [10] examines macroeconomic uncertainty's effect on Canadian banks, concluding that increased uncertainty inhibits loan scaling, which aligns with findings from Talavera et al. [11] regarding Ukrainian banks.

This research is applied in nature and falls under the category of correlational descriptive studies. The time frame for data collection spans the fiscal years from 2019 to 2023. In this study, considering the subject matter and its applicability, the research population includes all banks listed on the Tehran Stock Exchange that have been continuously active from 2019 to 2023. The banks under examination were determined using a systematic exclusion method, resulting in a total of 12 banks.

The statistical method used in this research is the GMMs. GMM is a robust estimation technique commonly applied in econometrics, particularly for dynamic panel data models. This method is advantageous because it does not require specific distributional assumptions about the error terms, making it ideal for handling endogeneity issues that often arise in economic models. In this paper, credit risk is introduced using the Non-Performing Loan Ratio (NPLR). The research equations include:

This research aims to investigate the relationship between EPU, credit risk, and lending decisions using the GMM method over the period from 2019 to 2023 for 12 banks listed on the Tehran Stock Exchange. The analysis of the three regression models provides significant insights into the dynamics of NPLs, LTDR, and ROA within the banking sector. Each model highlights different aspects of banking performance and risk management, contributing to a comprehensive understanding of how various factors interplay in influencing financial stability and profitability.

The results of Model (1) indicate a notable persistence in NPLs, as evidenced by the significant positive relationship between lagged NPLR and current NPLR. This suggests that banks with higher past NPLs are likely to experience continued challenges with NPLs, which can adversely impact their overall financial health. The significant negative relationship between Lev and NPLR implies that banks with higher Lev may adopt more effective risk management strategies, consequently reducing their exposure to NPLs. Furthermore, the positive correlation between bank size and NPLR raises concerns about the risks associated with larger institutions, potentially due to their complex operations and greater exposure to credit risk. In contrast, capital adequacy emerges as a critical factor; banks with higher capital ratios tend to have lower NPLs, reflecting stronger financial resilience.

The findings from Model (2) reveal that Lev and CARs significantly influence LTDR, suggesting that banks with higher Lev engage in more aggressive lending practices while maintaining adequate capital buffers. The marginally significant positive relationship between EPU and LTDR indicates that external uncertainties may slightly affect lending behaviors, although this effect is weak. The lack of significance for lagged LTDR and other variables suggests that more research is needed to identify additional factors influencing this ratio, as it plays a crucial role in assessing liquidity and operational efficiency within banks.

Model (3) demonstrates strong persistence in profitability, indicated by the significant positive relationship between lagged ROA and current ROA. This finding underscores the importance of historical performance in predicting future profitability. Additionally, the significant positive relationship between loan growth rates and ROA highlights the potential for banks that expand their lending activities to enhance profitability. While capital adequacy remains a vital determinant of ROA, indicating that well-capitalized banks are better positioned to achieve higher returns, other factors, such as EPU, do not appear to exert a meaningful influence on profitability.

In summary, these models collectively underscore the critical importance of effective risk management practices in banking. The persistence of NPLs poses ongoing challenges for financial institutions, particularly larger banks, that may face heightened risks. Furthermore, the relationships identified between Lev, capital adequacy, loan growth, and profitability highlight essential strategies for enhancing bank performance. Policymakers and bank management should prioritize improving capital buffers and monitoring lending practices to mitigate risks associated with NPLs while fostering sustainable growth in profitability. Future research should explore additional variables and external economic factors that could further elucidate the complexities of banking performance metrics.