دانشگاه اصفهان مدیریت دارایی و تامین مالی 2383-1189 14 3 2026 09 23 The Impact of Bank Capital on Liquidity Creation Across Quantiles: A Comparative Study of Developed and Developing Countries Using Quantile Regression Approach تأثیر سرمایه بانک بر خلق نقدینگی در دهک‌های مختلف: مطالعه تطبیقی کشورهای توسعه‌یافته و درحال‌توسعه با رویکرد رگرسیون کوانتایل 1 30 29704 10.22108/amf.2025.144564.1964 FA محمد سالم ماضی دانشجوی دکتری، گروه اقتصاد، دانشکده اقتصاد و مدیریت، دانشگاه ارومیه، ارومیه، ایران علی رضازاده دانشیار، گروه اقتصاد، دانشکده اقتصاد و مدیریت، دانشگاه ارومیه، ارومیه، ایران شهاب جهانگیری دانشیار، گروه اقتصاد، دانشکده اقتصاد و مدیریت، دانشگاه ارومیه، ارومیه، ایران رامین بشیر خداپرستی دانشیار، گروه امور مالی و بیمه، دانشکده اقتصاد و مدیریت، دانشگاه ارومیه، ارومیه، ایران Journal Article 2025 03 05  <br />This study investigates the nonlinear relationship between bank capital and liquidity creation across the distribution of liquidity creation (by deciles) in both developed and developing countries. Recognizing the critical role of liquidity creation in fostering financial stability and economic growth, the analysis addresses how the magnitude and direction of bank capital’s impact may vary across different levels of liquidity creation and country types. Employing quantile regression on a sample of 59 developing and 37 developed countries from 2004 to 2023, the findings reveal a consistently negative and significant effect of bank capital on liquidity creation throughout all deciles. However, this negative effect attenuates in higher deciles for developed countries, whereas it intensifies in developing countries. Furthermore, economic growth, financial inclusion, and financial development indices generally exhibit positive and significant effects. Conversely, the financial stability index demonstrates a significant negative impact in the lower deciles for developed nations and the higher deciles for developing economies. These contrasting outcomes underscore fundamental differences in liquidity creation mechanisms across countries and emphasize the necessity of a disaggregated, context-specific approach to banking regulation and policy formulation.<br /><strong>Keywords:</strong> Liquidity Creation, Bank Capital, Quantile Regression, Developing and Developed Countries<br /><strong>JEL Classification:</strong> G28, G21, C23<br /> <br /><strong>Introduction</strong><br />Liquidity creation, a fundamental function of banks within the modern financial system, involves the transformation of short-term liabilities into long-term loans, a process inherently coupled with risk intermediation. Within this framework, bank capital serves a dual and potentially contradictory role: while it provides a crucial buffer against losses that can enhance a bank's capacity to assume risk and create liquidity, it may also constrain the resources available for lending. This tension suggests a nonlinear relationship, wherein the effect of capital on liquidity creation may vary across its distribution. At lower capital levels, increases might initially curb liquidity creation by reducing lendable funds, whereas at higher levels, the enhanced risk-absorbing capacity could facilitate greater liquidity creation.<br />This non-uniformity implies that the impact of capital regulations is likely heterogeneous across institutions, challenging the efficacy of a one-size-fits-all regulatory approach. Motivated by this complexity, the present study employs a quantile regression (QR) methodology to empirically investigate the nuanced relationship between bank capital and liquidity creation across different deciles of the liquidity creation distribution. By analyzing a global sample of 37 developed and 59 developing countries over the period 2004–2023 within separate models, this research aims to provide a more disaggregated understanding critical for designing targeted prudential policies.<br /><strong> </strong><br /><strong>Materials & Methods</strong><br />This study employs a panel quantile regression (QR) methodology to examine the nuanced impact of bank capital on liquidity creation across the entire conditional distribution of the latter. Selected for its capacity to provide a comprehensive analysis beyond the conditional mean, this approach is particularly advantageous for capturing potential heterogeneity in the relationship across different deciles, including the tails of the distribution. In contrast to conventional ordinary least squares (OLS) regression, QR estimates coefficients by minimizing a weighted sum of absolute deviations, known as the Least Absolute Deviation (LAD) method. This technique is robust to non-normal error distributions, heteroscedasticity, and the presence of outliers, thereby yielding more reliable and efficient estimates for our financial dataset, which may exhibit such characteristics.<br />Aligned with this rationale, we estimate a nonlinear panel quantile regression model for a global sample comprising 37 developed and 59 developing countries over the period 2004–2023. The general empirical specification, adapted from the frameworks established by MazioudChaabouni et al. (2018) and Gupta et al. (2023), is formally defined as follows:<br /><br /><br /><br /><br /> <br /><br /><br />(1)<br /><br /><br /><br /><br />in which, <strong>LM</strong> represents liquidity creation, <strong>LCAR</strong> represents bank regulatory capital (transition variable), <strong>LBZSCORE is the</strong> financial stability, <strong>LGDPP</strong> is the variable of economic growth, <strong>LATM</strong> represents the financial inclusion index, and <strong>LFSD is the</strong> financial development index.<br /><strong> </strong><br /><strong>Findings</strong><br />The stationarity of the variables was assessed using the Levin, Lin, and Chu (LLC) unit root test. As detailed in Table 1, the results confirm that all variables are stationary at the 5% significance level, incorporating an intercept term. Subsequently, the core empirical analysis, illustrated in Figure 5-a, reveals a statistically significant negative relationship between bank capital (LCAR) and liquidity creation (LM) across all deciles for developed countries. Notably, the magnitude of this negative effect exhibits a diminishing pattern, weakening progressively throughout the higher deciles of the liquidity creation distribution.<br /> <br /> <br /><strong>Chart (5-a). Trends in variables in deciles in developed countries</strong><br /> <br />Furthermore, the results for developing countries, as visualized in Figure 5-b, indicates a distinct and evolving relationship. The effect of bank capital (LCAR) on liquidity creation (LM) is statistically insignificant and positive in the first decile. However, this relationship transitions to a negative and statistically significant influence beginning in the second decile. Moreover, the magnitude of this adverse effect demonstrates a pronounced intensification across the higher deciles of the distribution.<br /> <br /> <br /><strong>Chart (5-b). Trends in variables in deciles in developing countries</strong><br /><strong> </strong><br /><strong>Conclusion and discussion</strong><br />In conclusion, this study establishes that the relationship between bank capital and liquidity creation is not only nonlinear but also contingent upon a country's developmental context and its position within the liquidity creation distribution. The analysis reveals a consistently negative yet diminishing effect across deciles for developed nations, while in developing countries, the relationship manifests as negative and significant from the second decile onward, intensifying markedly at higher levels. This stark heterogeneity underscores fundamental differences in the operational mechanisms of financial intermediation and the distinct role capital plays across diverse economic landscapes. Furthermore, control variables corroborate this complexity; economic growth, financial inclusion, and financial development predominantly exert a positive influence on liquidity creation, whereas financial stability exhibits a significant negative impact in specific deciles, particularly within developing economies. Collectively, these findings carry substantial policy implications, strongly advocating for a disaggregated regulatory approach.<br />Policymakers must therefore eschew uniform, one-size-fits-all capital regulations in favor of frameworks meticulously tailored to a country's level of financial development and the specific characteristics of its banking institutions. The application of quantile regression in this analysis proves indispensable, providing the granular insights necessary for such precise and effective policy formulation. مسئلۀ اصلی این پژوهش، بررسی رابطۀ غیرخطی میان سرمایۀ بانکی و خلق نقدینگی در سطوح مختلف توزیع خلق نقدینگی (دهک‌ها) در کشورهای توسعه‌یافته و درحال‌توسعه است. اهمیت این موضوع از آنجا ناشی می‌شود که خلق نقدینگی نقش کلیدی در ثبات مالی و رشد اقتصادی ایفا می‌کند؛ اما میزان و جهت اثرگذاری سرمایۀ بانکی ممکن است در سطوح مختلف خلق نقدینگی و بین کشورها متفاوت باشد؛ در این راستا، مطالعۀ حاضر با استفاده از رگرسیون کوانتایل، اثر سرمایۀ بانکی را بر خلق نقدینگی در دو مدل جداگانه برای 59 کشور درحال‌توسعه و 37 کشور توسعه‌یافته طی دورۀ 2004 تا 2023 بررسی می‌کند. یافته‌ها نشان می‌دهد که سرمایۀ بانکی در تمامی دهک‌ها اثری منفی و معنادار بر خلق نقدینگی دارد، به‌گونه‌ای که این اثر در دهک‌های بالای کشورهای توسعه‌یافته کاهش یافته، ولی در کشورهای درحال‌توسعه افزایش یافته است. همچنین، رشد اقتصادی، شاخص شمول مالی و توسعۀ مالی عمدتاً اثر مثبت و معنادار دارند، درحالی‌که شاخص ثبات مالی در دهک‌های پایین کشورهای توسعه‌یافته و دهک‌های بالای کشورهای درحال‌توسعه اثر منفی معناداری دارد. این نتایج گویای تفاوت در سازوکار خلق نقدینگی بین کشورها و اهمیت رویکرد تفکیکی در سیاست‌گذاری است.<br />  https://amf.ui.ac.ir/article_29704_a231a48bcff15900e8f9ceac00b84196.pdf