Bank Lending and Policy Interactions - A Comprehensive Assessment for the G20 Countries
Abstract
EconStor is a publication server for scholarly economic literature, provided as a non-commercial public service by the ZBW.
Full text
Gramlich, Dieter; Yan, Meilan; Zhang, Dalu Article — Published Version Bank Lending and Policy Interactions - A Comprehensive Assessment for the G20 Countries International Journal of Finance & Economics Provided in Cooperation with: John Wiley & Sons Suggested Citation: Gramlich, Dieter; Yan, Meilan; Zhang, Dalu (2024) : Bank Lending and Policy Interactions - A Comprehensive Assessment for the G20 Countries, International Journal of Finance & Economics, ISSN 1099-1158, John Wiley & Sons, Ltd., Chichester, UK, Vol. 30, Iss. 4, pp. 3500-3520, https://doi.org/10.1002/ijfe.3075 This Version is available at: https://hdl.handle.net/10419/329816 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/4.0/
International Journal of Finance & Economics, 2025; 30:3500–3520 https://doi.org/10.1002/ijfe.3075 3500 International Journal of Finance & Economics RESEARCH ARTICLE OPEN ACCESS Bank Lending and Policy Interactions—A Comprehensive Assessment for the G20 Countries DieterGramlich1 | MeilanYan2 | DaluZhang3 1DHBW Baden- Wuerttemberg Cooperative State University, Heidenheim, Germany | 2Loughborough Business School, Loughborough University, Loughborough, UK | 3School of Business, The University of Leicester, Leicester, UK Correspondence: Dieter Gramlich ([email protected]) Received: 20 April 2021 | Revised: 24 June 2024 | Accepted: 31 October 2024 Funding: The authors received no specific funding for this work. Keywords: financial regulation| interaction| monetary policy| rescue package ABSTRACT This study examines the joint impact from supervisory requirements, monetary policies and rescue packages on the supply of bank loans. Evidence is obtained from conceptual considerations and the empirical investigation of G20 banks over the period 1995–2021. To prepare the analysis of interactions, we first address modelling concepts and consistently identify the effects on bank lending that come from each of the regulatory approaches in isolation. Second, we empirically assess the effects that result from the parallel application of policy actions. S hortterm liquidity ratios in combination with further variables mostly associate with positive interaction effects. In contrast, the joint assessment of longterm liquidity ratios as well as leverage ratios with other policy actions usually provides negative effects. Overall, the results suggest that unconventional monetary policy and rescue actions can stimulate bank lending only after the institutions have restored their own stability. Interactions of policy actions matter, and both regulatory authorities and bank management should consider them. 1 | Introduction In recent years, the regulatory framework for the banking business has evolved steadily (BCBS 2022; Evgenidis and Papadamou 2021). While supervisors of the financial system and monetary policy makers are generally constantly reassessing exposures in a dynamic financial environment, the subprime crisis, the subsequent European debt crisis and the COVID- 19 pandemic, in particular, have fueled the increase in regulatory policies (Cortes etal.2022). This is manifested in a series of worldwide reforms in banking supervision, the Basel I–IV accords (supervisory policy), a shift in central bank policy (monetary policy) as well as in multiple interventions and rules for rescuing distressed banks (rescue policy).1 We refer to the entity of these institutional settings as the totality of policy (EBA2015; Llewellyn2015). Figure1 provides an overview of the evolution of the banking policy framework. The Basel I–IV accords aim to improve both the resilience of individual banks to adverse effects on solvency and liquidity as well as the overall stability of the banking system (BCBS2022; Jabbour2022). While Basel I, which came into force in 1988, implemented a simpler and more schematic relationship between the credit business of banks and their capital, the Basel II reform of 2006 addressed in much more detail the measurement of default risk and added market risk as well as operational risk to the risks that banks' capital must cover. The subsequent framework, Basel III, finalised in 2011, has narrowed the components of regulatory capital and increased capital ratios (BCBS2011). It introduced the net stable funding ratio (NSFR) to align longterm investments with longterm funding, and the liquidity coverage ratio (LCR) to address the banks' resilience to shortterm liquidity shocks. The newly introduced leverage ratio is a purely volumebased ratio and aims at limiting the sheer size of banks' business. Major changes during the final implementation This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made. © 2024 The Author(s). International Journal of Finance & Economics published by John Wiley & Sons Ltd.
3501 of Basel III, in particular regarding the use of internal models and the weighting of counterparty risk, more recently mark the transition to Basel IV (Feridun and Ozun2020). Against the backdrop of the increasing fragility of the financial system, central banks took unprecedented monetary actions (Avalos and Mamatzakis2023; Martins, Batista, and Ferreira- Lopes2019). Within the policy of quantitative easing and as a specific response to the financial turmoil, they provided generous shortterm liquidity to banks, expanded the direct purchase of securities from banks and significantly lowered interest rates (Gern etal.2015). For example, the monetary policy operations of the European Central Bank (ECB), as represented in the consolidated balance sheet of the Eurosystem, increased from €269bn at the end of 2000 to €682bn at the end of 2010 and €5488bn at the end of 2020 (ECB2022). In addition to prudential and monetary actions aiming at supporting the banking system as a whole, policy makers have implemented special bailout packages for individual distressed institutions. Thereby, governments together with monetary authorities injected vast amounts of capital, purchased banks' nonperforming assets, provided guarantees and increased the coverage of deposit insurance schemes (Allen etal.2015; Palas and Moreira 2022). Further to providing liquidity and taking stakes in banks, they have defined the criteria under which rescue actions should be carried out.2 The different policies have contributed to the successful management of periods of financial market distress and to a more resilient banking system. When the individual actions are considered, they are commented positively (Laeven, Maddaloni, and Mendicino2022; Martin, Mendicino, and Van der Ghote2022). However, it is also critically asked if the effects from the actions were optimal when considered in their totality, that is, seen from their joint effects (Laeven, Maddaloni, and Mendicino 2022) and ‘as opposed to each regulation taken alone’ (EBA2015, 20). Regulatory policies are set by different institutions and have individual objectives (Buch etal.2021; Claessens etal.2013), and their parallel implementation may have unintended cumulative (amplifying) or compensating (offsetting) effects. As the multiple actions affect banks simultaneously, their aggregated outcome may differ from the sum of the outcomes of single actions. A major concern is that the ability of banks to provide sufficient credit to the economy is obstructed (Cecchetti2015; Mesonnier and Monks2015). Llewellyn(2015) emphasises that examining the totality of policy performance is very much essential for policy makers to understand fully the effects of regulation and regulatory policies. There is a symbiotic relationship between regulation and banking behaviour, where both respond to each other. Regulation eventually does not only affect the behaviour of those banks specifically targeted, but these behavioural changes would also lead to feedbacks to the system in general. Further to looking at the single actions' impact in isolation (individual effect), the question arises what has been their additional effect in the context of further policy actions (interaction effect), that is, the ‘marginal cost and marginal benefit of regulation’ (Llewellyn 2015, 66). Similarly, Vo(2021) argues from the perspective of prudential regulation that knowledge about the interaction between capital and liquidity requirements is essential to assess the efficiency of the postcrisis framework. Interactions between regulatory policies may further affect the discussion about the institutional setting of regulation, that is, the question if prudential and monetary policies should be conducted from different institutions or ‘under one roof’ (Malovana etal.2023, 1). In case there are major interactions, regulatory policies should better be overseen from one institution, where in the case of weak or no interactions institutions may act on their own. Although the topic of policy interactions has recently attracted interest, evidence about the interplay of regulatory actions remains scarce. The overview study from Bussiere etal. (2021) documents current work at six central banks and the Bank for International Settlements (BIS). The authors conclude that the ‘interaction of monetary and macroprudential policy, however, remains underexplored, both academically and from a policy perspective’ (Bussiere etal.2021, 2). In their review of studies on the interaction between macroprudential and monetary policy mainly conducted at the ECB, (Martin, Mendicino, and Van der Ghote2022, 1) note that the ‘quantification of this tradeoff remains an exciting question’. The BCBS(2022) report comes to a similar conclusion. FIGURE 1 | Timeline of supervisory, monetary and rescue policies. Greecebailout (2010) ´88´97 ´99´01 ´03´05 ´07´09 ´11´13 ´15´17 Basel III (finalised 2011, effective 2014) -increased capital -leverage ratio -LCR -NSFR Basel II (finalised 2004, effective 2006) -risk quantification -internal models Basel I (effective 1988) -risk assessment -capital ratio New economy bubble (2000) Quantitative easing Japan (since 2001) TARP (2008-14) Dodd-Frank (2010) EFSF (2010) SMP (2010) Quantitative easing USA (2008-14) Quantitative easing EU (2015) MREL and TLAC forG-SIBs (2015) Lehman Brothers (2008) LTCM collapse (1998) ´19´21 Capital Requirements Regulation III (Basel IV, 2021) Pandemic emergency purchase program (2020)
3502 International Journal of Finance & Economics, 2025 Considering the findings from these studies and further literature (see Section2), there is a need to further examine the different policy measures from a holistic perspective and to assess their integrated effects. To combine the different policy measures in a coherent way, we must first establish a common basis for making the individual policies comparable. Our first research question is therefore: 1. What is an appropriate conceptual framework for consistently assessing the effects of policy interactions? Once we have established this unifying framework, we can further analyse the quantitative effects that result from the joint application of policies. In particular, we look at the effects of the set of policies on banks' supply of money to the economy. Our second research question is: 2. What is the overall quantitative impact of policy interactions on bank lending? While taking a further step in the analysis of regulatory policy interactions, our study extends knowledge in two ways: We first contribute to conceptual considerations suggesting a classification of the different types of interactions within the totality of policy and discussing potential channels of interaction effects. We also aim to identify the effects from different single policy actions in a consistent manner, which forms the basis to further quantify the joint effects of multiple policy actions. Our second contribution is to empirically assess the linkages between simultaneously applied policy actions. We examine the interaction (marginal) effects on bank lending that arise from the parallel application of single supervisory, monetary and rescue policies, thereby emphasising potential reinforcing or offsetting effects from their interplay. Depending on the number of single policy actions included in our integrated analysis, we distinguish 2- type interactions (two different policies are considered together), 3- type interactions (three different single actions are combined) and higher types of interactions. Our results may not only support policy makers in efficiently coordinating and jointly applying their strategies (Pierret2015), but also may contribute to the institutions´ individual risk management in showing the need to focus risk management activities in the context of multiple regulatory requirements. Our empirical investigation focuses on the interaction effects on bank lending as a measure of banks' dynamic and monetary supply to the economy. We include 1052 banks from 20 major countries worldwide (G20) where all banks provide a complete set of annual data during the period 1995–2021. Using both panel fixed effect (panel- FE) estimator and the system generalised method of moments (S- GMM) estimator to relate the different supervisory, monetary and rescue actions to bank lending, we obtain consistent results for joint policy effects from the product terms of regression coefficients. Among the main results, we find that with respect to the interplay between two policies (referred to as a 2- type interaction), capital and liquidity requirements have a positive interaction. This important result suggests that supervisory authorities can use these measures together without compromising their overall objectives. We also find that all 2- type interactions that include the leverage ratio, have negative coefficients. Based on this finding and the negative single effect, we conclude that the leverage ratio dominates other regulatory actions, and policymakers should be mindful of the strong and overproportional impact from this ratio. We further examine higher types of interactions and finally include all supervisory, monetary and rescue actions (totality of policy interactions). However, the results become less significant and are difficult to interpret. The evidence from higher types of interactions thus remains limited. Our study is structured to follow through with next reviewing the findings from literature on the interaction effects of regulatory policies, notably on bank lending. We use this review to develop conceptual considerations for the analysis of interaction effects (Section2). We establish our hypotheses, data and methodological framework with a special focus on the interaction effects and choose a GMM approach (Section3). Section4 provides main findings and particularly discusses the marginal effects from simultaneous regulatory actions, and Section5 concludes. 2 | Evidence From the Literature While the literature offers several discussions about the impact from single regulatory policies on bank lending, research on the effects from policy interactions ‘is more recent, and the literature is still at the early stage’ (Vo2021, 21). We provide an overview of both directions of research. Considering the findings from literature and the scarce evidence about policy interactions and bank lending, we suggest a classification of interaction types and discuss channels of interaction effects as a basis to further investigate them empirically. 2.1 | Research on Single and Interaction Effects Theoretical discussions emphasise the contradictory effects, which tough policies of bank capital and liquidity can have on bank lending (e.g., Peek and Rosengren 1995; Peura and Keppo2006). Authors argue on the one hand that in order to achieve the minimum Basel capital and liquidity ratios, banks become reluctant to approve new lending, begin to collect outstanding loans or offer more attractive long term bonds. On the other hand, banks with a high capital ratio may be considered to have a strong basis and provide a higher credit supply. In their empirical study, Gambacorta and Marques- Ibanez(2011) show that banks with weaker core capital positions, greater dependence on market funding, and on noninterest sources of income restricted loan supply more strongly during the crisis period. Buch and Prieto(2014) find from an investigation of bank groups in Germany over 44 years that in the long run higher bank capital is associated with a higher volume of business loans. Mesonnier and Monks(2015) conclude from an analysis of around 200 large European banks for the period 2011–2012 that banks constrained to strengthen their core capital ratio displayed a lower growth in credit than those without additional capital requirements. With respect to large banks in the United States, Kim and Sohn(2017) find that a higher level of bank capital only contributes to credit growth when the banks also retain sufficient liquid assets.
3503 The bank lending channel of monetary policy suggests that banks play a special role in the transmission of monetary policy (Martins, Batista, and Ferreira- Lopes 2019). In this theory, monetary policy has both an effect on the change in the riskfree rate and on banks' funding costs leading to an additional response in bank lending (Black, Hancock, and Passmore 2010). However, there is no consistent empirical evidence. Kashyap and Stein (1995) suggest that expanding monetary policy leads to an increase in lending supply, which varies for large and small banks. Kishan and Opiela(2000) suggest that banks might restrain lending following an increase in the federal funds rate if they face liquidity constraints or low capital levels. Among the few studies for the European market, Martins, Batista, and Ferreira- Lopes (2019) find that unconventional monetary policy positively affects bank credit, whereby lending to governments dominates lending to households. An alternative view of the bank lending channel focuses on the extent the monetary transmission mechanism impacts changes in banks' business models and market funding patterns (Borio and Disyatat2011; Gambacorta and Marques- Ibanez2011). In addition, the impact of rescue policies on bank lending has been subject of longstanding debates, mainly because of the magnitude of moral hazard costs. Already Bagehot(1873) argues that support to bad banks would cause even worse lending decisions. Diamond and Rajan(2000) and Diamond(2001) point out that too small recapitalization may even be damaging for bank lending policies. A group of studies highlight that recapitalization has to be sufficiently large to solve banks debt problems before increasing bank lending (e.g., Bhattacharya and Nyborg2011; Giannetti and Simonov2012). Brei, Gambacorta, and von Peter(2013) show that a stronger capitalization sustains loan growth in normal times, whereas during a crisis, banks can turn additional capital into greater lending only once their capitalization exceeds a critical threshold. This suggests that recapitalization may not translate into greater credit supply until bank balance sheets are sufficiently strengthened (Bernanke and Lown1991). Whereas already the evidence from single policy approaches is not fully obvious, a further challenge comes from the parallel application of these policies. As has been mentioned, a series of regulatory actions with different objectives, policy makers and timing might cause unexpected consequences due to interactions among the policies. The scarcity of evidence about policy interaction effects may be because single policies had to be quickly implemented during crises where not much time had been left to fully consider their interconnectivity. Also, Basel III as well as monetary actions during the pandemic have been implemented more recently, and the number of observations is sparse as well (Vo2021). The ‘very limited number of empirical studies’ (Birn, Dietsch, and Durant2017, 11) is further explained by the fact that already the single actions within a regulatory direction are complex. The complexity increases within the totality of interactions. Schmaltz etal.(2014) investigate the joint effects from the four Basel constraints based on a model of the banking firm and apply the model to accounting data for an average bank. They optimise the banks' profit subject to the capital, leverage and liquidity constraints and find that banks would increase liquidity reserves at the expense of lending to private customers. Based on an accounting model for banks, Birn, Dietsch, and Durant(2017) assess the distance to compliance with Basel ratios. They further develop an adaptation strategy and verify it based on empirical bank data from quantitative impact studies. A main result is that while both increasing capital ratios and liquid assets, the observed banks extend their lending to nonfinancial sectors. DeYoung, Distinguin, and Tarazi(2018) study how banks adjust both their liquidity and capital management following an unexpected, exogenous shock of the capital ratio. They find for community banks in the US that smaller banks reacted to involuntary reductions of their capital ratio by increasing capital and liquidity ratios in parallel, however, at the expense of loans and loan commitments. The authors could not find a similar effect for large banks. Authors from the ECB examine the connectivity between different macroprudential and monetary measures and its impact on the effectiveness of regulatory policies. Cozzi et al. (2020) identify a kind of competition between macroprudential and monetary actions. As regards the effects from capital requirements, they state that the more monetary policy leans against the burden of higher capital ratios, the smaller is the decline of lending and the output of the economy. Based on the results from four general equilibrium models, they further find that monetary policy is less powerful when banks' leverage ratio is high whereas lower leverage ratio increases the economy's sensitivity to monetary actions. Hoerova etal.(2018) examine the aggregate effect of liquidity ratios in combination with capital requirements and actions from a lender of last resort on the solvency and liquidity of banks. They obtain results based on the application of macrofinancial models and data from European banks' balance sheets. They find that with respect to bank liquidity, liquidity requirements serve as a substitute for capital requirements, and that higher liquidity requirements in the past would have reduced the extent of actions from the lender of last resort. The comprehensive BCBS (2016, 2022) surveys of literature as well as the overviews on policy interactions from Bussiere etal.(2021), Martin, Mendicino, and Van der Ghote(2022) and Vo (2021) all state that the joint effects from the implementation of supervisory, monetary and rescue policies matter, but are scarcely researched up to now. Vo(2021) particularly highlights the importance of further research of the effects of policy interactions on the cost and volume of lending. Building on these studies, the authors suggest further theoretical work and evidence from empirical investigations to explore the various dimensions of policy interactions. These dimensions include investigating propagation channels through which policy measures affect lending, analysing the systemic effects resulting from the interactions, and assessing the overall impact on bank lending. 2.2 | Assessment of Policy Interactions Interaction effects occur because of the different objectives pursued with regulatory policies and the subsequent reaction of banks to these policies. They originate from the decisions
3504 International Journal of Finance & Economics, 2025 about single regulatory policies, involve connectivity during the transmission of the policies and jointly affect the output on the targeted variable. Regarding the parallel application of single regulatory policies as the origin of connected effects, we distinguish two levels of interactions as follows (see Figure2). First of all, interactions within one of the three basic approaches of policy refer to the connectivity of single policy actions inside each supervisory, monetary and rescue policy. The Basel supervisory framework itself includes four basic options, that is, a capital ratio (where both riskweighted assets and recognised capital may change), a leverage ratio, a shortterm and a long - term liquidity ratio (BCBS2011). A first issue is therefore to investigate what effects come from the joint application of these requirements from banking supervision (this is referred to as supervisory interaction). Hereby, interactions between two, three or four options may be considered. Similarly, the policy of quantitative easing which itself grounds on the volume and the price of money (monetary interaction) and rescue packages must be considered that include capital injections, liquidity provisioning and deposit insurance (rescue interaction). Second, interactions between policy approaches include the connectivity among the three basic policies. This is referred to as totality of interaction and comprises the interplay of single actions from the three regulatory policies. Regulatory policies connect via different transmission channels where they affect the involved institutions and parameters in a similar or different way. Claessens etal.(2013) suggest a risktaking channel and an assetprice channel to explain policy interactions. The risktaking pathway includes the change in bank leverage as well as in the tightness of borrowing constraints to adjust to policy requirements. In their view, prudential requirements contribute to moderate financial (lending) stress in times where unconventional monetary policies are applied. As concerns the assetprice channel, monetary policies, for example, a decrease in interest rates, trigger the value of assets with a further impact on collateral used for lending. The BCBS(2016) suggests four interaction channels including the quality of assets, firesales, bank profitability and bank solvency. Policy actions propagate through these channels producing multiple results. Regarding liquidity ratios as an example, the authors see a similar effect from the NSFR and LCR where for both ratios the banks attempt to increase the quality of their assets. The switch to lowrisk and marketable securities mainly involves a negative effect on bank lending. Bussiere etal.(2021) refer to the bank lending channel and the portfolio channel (balance sheet channel) as the main transmission pathways. The cost of raising money (interest rate policy) and liquidity restrictions (prudential policy) affect the actions of banks in the lending channel, whereas the portfolio channel includes changes in the structure of assets due to changes in the risk perception of institutions. Summarising the literature, we suggest four main channels where regulatory policies propagate and mutually connect with a subsequent effect on bank lending. At the same time, these channels are interrelated. Emphasising the transmission mechanisms rather than the transmission incentives, interactions emerge from first a value channel where banks adjust to regulatory actions constraining the relation between assets and equity. Within, second, the liquidity channel, policies have an impact on lending because regulatory actions require adjustments to their cash position. Within, third, the risk channel, regulatory policies affect the risk sensitivity of banks and their willingness to further expand or contract the lending business. Fourth, a further channel may include additional connections not yet captured by the first three. For example, a further interaction effect is that simply the extent of policy requirements may overburden the institutions and have negative effects on their ability to provide financial services (BCBS2016; Krahnen, Noth, and Schüwer2017). The parallel increase of supervisory and monetary actions restricts the flexibility of banks with a further impact on their lending. Finally, the overall outcome of policy interactions is that they alter the desired results from single policies. The effect from multiple actions is not just the sum of single effects but can be lower or even higher. Following the BCBS(2016) and Vo(2021), the joint effects may be complementary, where two or more policy actions have an additional impact on the targeted variable (marginal effect is positive, integrated effect is larger than the sum of individual effects) or may be substitutional, where the parallel implementation of policy actions does not yield any increase in output, rather the policies are overlapping or even offsetting each other (marginal effect is zero or negative). Van der Ghote(2021) models the effects that the combination of prudential and monetary policies can have over a boombust cycle, that is, pointing to the lagged effects of interactions and their combined assessment across time. FIGURE 2 | Policy interactions. Totalityof interaction Rescue policy Supervisory policy Monetary policy -Higher liquidity -Lower interestrate -Capital ratio -Short-term liquidity -Long-term liquidity - Leverageratio -Capital injection -Liquidity provision -Deposit insurance Interaction betweenpolicies Interaction within policies
3505 For example, a complementary or cumulative interaction effect can stem from prudential ratios. Limits to the volume of business imposed from the leverage ratio in conjunction with the unfavourable inclusion of loans within the calculation of LCR and NSFR may have an amplifying impact and create a cumulative incentive to cut lending. As an example for substitutional or compensating effects, the banks may consider credit rationing to comply with the increasing scale of capital requirements in the Basel accords. Concurrently, they may also tend to increase credit supply as they receive generous funding from nontraditional monetary policy (notably, the purchase of assets) and capital injections from government liquidity schemes (Hoerova etal.2018). Cappelletti etal.(2022) find a tradeoff from targeted longerterm refinancing operations (TLTROs). Banks engaged in TLTROs restrict their lending to other banks but not to nonfinancial corporations (TLTROs link borrowing from the central bank to lending to nonfinancial corporations). In summary, policy interactions must be seen as complex, nonlinear and changing phenomena. Further to the different objectives, parameters and pathways involved in their propagation, the specific context of the institutions and the markets will have an impact on their overall effects. Combinations of specific regulatory policies will result in different outcomes, and there ‘is not a onesize- fitsall channel or even direction of transmission’ (Bussiere etal.2021, 4). Therefore, we next develop several expectations for the joint outcome of policies depending on the individual setting of policy actions (Section3.1) and test these hypotheses individually in the empirical part of our study. We present the data basis for the empirical setting (Section3.2) and further develop our econometric specification (Section3.3). 3 | Methodology 3.1 | Hypotheses Development We develop our hypotheses by following and extending previous findings from the literature. Table1 provides an overview of methodologies and results applied in the literature for the analysis of regulatory interactions. The total effect from a specific regulatory policy on lending includes the direct or single effect from this policy variable plus the effect from the variable's connectedness with other regulatory actions. Schmaltz etal.(2014) model how a typical German bank can maximise the accounting profit under Basel III ratios. In their profit optimization model, they define the Basel III ratios as four linear constraints. However, the effects from their connectivity remain implicit as the constraints simultaneously impact the optimal business structure, which is found as the tradeoff between margins and costs from adjusting bank business to the Basel requirements. The authors do not explicitly model the effects from the inclusion or exclusion of single constraints, rather they emphasise the effects from changing margins and adjustment costs. Similarly, Birn, Dietsch, and Durant(2017) develop the optimal profitcost equation by considering possible changes in bank business models. Their approach takes explicitly into account the restrictions the new regulations impose on banks' balance sheets as well as the mutual interactions between new regulatory capital and liquidity constraints. Brei, Gambacorta, and von Peter(2013) and Kim and Sohn(2017) already apply linear regression to investigate the effects from bank capitalisation as a single regulatory policy. We partly follow their methodology. However, as our main objective is to assess the aggregate effects from multiple potentially complementing or conflicting policies, we particularly develop interaction terms to be included in the regression model. A common way to account for the connectivity between two explanatory variables in regression is to multiply the variables and include the product term in the econometric equation (Jaccard and Turrisi 2003; Wooldridge 2021). Through multiplication, the joint effect from the variables is expressed to either strengthen, lower or neutralise the effects from single variables. The sign and size of the product coefficient show, if and how much the respondent variable is specifically impacted from considering the explanatory variables simultaneously. In a more recent and preliminary approach, the BCBS(2022) similarly applies product terms to investigate interactions. We relate our results to existing (few) research and present additional findings for the totality of regulation. Our expected signs of the impact on bank lending of changes in response to different policy interactions are summarised in Table2. 3.2 | Construction of the Dataset We include all the G20 banks from our database, which provide a complete set of data for the overall period. Our sample covers 1052 active banks' consolidated annual data from 1995 to 2013 in Bankscope (1995–2013) (Bureau van Dijk2017) and Fitch Connect (2014–2021) (Fitch Solution2023)3 to capture the whole Basel framework which includes the first draft of Basel I released in 1988, Basel II in 2004, Basel III finalised in 2011 and the modifications during its implementation (Basel IV). The G20 countries are chosen as they comprise the majority of banks affected from the different regulatory policies, and as the G20 has incentivised many of the regulatory reforms in banking and finance (Bundesbank2016). The choice of consolidated bank financial statements is in line with the practical view that banks take decisions on their consolidated assets and liabilities. It further considers that rescue measures are provided to the consolidated entity rather than to individual subsidiaries by the host countries. The macroeconomic variables are collected from Datastream. For bank rescue information, we hand collect each G20 country central bank's bailout scheme for the years 1995 to 2021. The information regarding the years affected by the COVID- 19 pandemic in each G20 country is sourced from the World Health Organisation. As the 27- year data sample comprises not only different economic cycles but also different organisational transformations of individual banks, it is essential to check the mergers and acquisitions (M&A) history of each bank over the sample period. Brei, Gambacorta, and von Peter(2013) argue that the control of banks undergoing mergers could help to exclude spurious bursts of lending growth. Therefore, we adjust for 435 mergers and acquisitions by separating target bank and acquiring bank into two different consolidated groups based on the M&A date from Bureau van Dijk's Zephyr database. Table3 shows all the active banks that are headquartered in G20 countries, along with their 27- year average assets, net loans, and equity levels.
3506 International Journal of Finance & Economics, 2025 The main variables in this study are chosen with respect to the bank lending channel literature (see Table4). Following most of the studies (e.g., Berrospide and Edge2010; Brei, Gambacorta, and von Peter2013; Gambacorta and Mistrulli2004; Kashyap and Stein1995; Kim and Sohn2017), we measure bank lending as the growth of bank net loans to nonfinancial institutions ( ΔL ). However, we recognise that banks provide credit to the real economy by other means as well, for example, when they purchase bonds from companies.4 This study takes into account the most relevant requirements from the Basel framework. As a representation of the capital requirements, we choose Core Tier 1 Regulatory Capital Ratio ( TIE ). This ratio based on Core Tier 1 capital is considered to be more challenging than other capital ratios as it focuses on the level of highquality capital.5 We expect the coefficients for the Tier 1 capital ratio to be positive because well capitalised banks can more effectively absorb the negative effects of shocks on bank lending (Kim and Sohn2017). TABLE 1 | Interactions of regulatory polices and bank lending—evidence from literature. Author(s) Topic Methodology Data Findings (impact of variable(s) on loans) Brei, Gambacorta, and von Peter(2013) Rescue packages in crisis periods GMM panel 1995–2010 CAP ×RES(+) (recapitalization) Rescue dummy 14 countries Only if CAP >threshold Covas and Driscoll(2015) Basel III capital and liquidity requirements Dynamic general equilibrium model 1997–2012 CAP ×LIQ(−) For calibration CAP ×LIQ(+) Schmaltz etal.(2014)Adjustment effects from Basel III constraints Optimization model Average CAP ×LIQ ×LEV(−) Bank GER Model approach De Nicolo, Gamba, and Lucchetta(2014) Basel III capital and liquidity requirements Equilibrium framework 1983–2009 CAP ×LIQ(+) Birn, Dietsch, and Durant(2017) Effect of Basel III ratios: bank assets and deposits Optimization model, nonlinear 2011–2014 CAP ×LIQ(+) 156 banks Interbank loans (+) Kim and Sohn(2017) Bank capital and liquid assets Fixed effects panel, interaction 1993–2010 CAP ×LIQ(+) US relationship not linear DeYoung, Distinguin, and Tarazi(2018) Capital shock and small banks' liquidity behaviour Model exogenous shock and reaction 1998–2010 CAP ×LIQ(−) US indirect evidence Carletti, Goldstein, and Leonello(2019) Impact of capital and liquidity on bank stability Equilibrium model, balance sheet constraints CAP ×LIQ(−) BCBS(2022)Interaction of capital and liquidity on lending Fixed effects panel, interaction 2012–2021 RegRatio1 ×RegRatio2 (±) 34 banks Note: CAP is both TIER 1 and TIER 2 capital over risk weighted assets, LIQ is liquidity assets to total assets, RegRatio represents Common Equity TIER 1, LCR, NSFR, Leverage ratio.
3507 The leverage ratio ( LEV ) is defined as the relation between Tier 1 capital and the banks' average total consolidated assets (sum of the exposures from onbalance and offbalance sheet items). It is also called a nonrisk- based ratio as the onbalance and offbalance items are included based only on their nominal values. Particularly, it is intended to restrict the buildup of excessive on- and offbalance sheet business volumes in the banking system to avoid destabilising deleveraging processes which further can damage the broader economy (Feridun and Ozun2020). The nonrisk- based leverage ratio is therefore an independent measure and complements the risk capital requirements. We expect that a higher leverage ratio affects the ability of banks to maintain loan growth and to raise funds (Kishan and Opiela2000). In line with Gambacorta and Marques- Ibanez(2011), we also incorporate the banks' deposit to total funding ratio ( DEP ) to account for the increased lending capacity of banks with more stable deposit funding. However, considering the high correlation observed between the variables LEV and DEP , we only keep LEV in our estimation process to mitigate potential multicollinearity concerns. With respect to LCR and NSFR, current databases do not yet contain sufficient information to directly assess the newly TABLE 2 | Effects from interactions—Expectations for the supply of loans. Policy interactions Expected signs Basic arguments & discussions Supervisory & Supervisory Capital × Long term liquidity + Well capitalised banks may also better match long term funding requirements what tentatively supports a positive interaction. Capital × Short term liquidity + Highquality liquid assets (HQLA) usually have no risk and higher capital may not compromise short term liquidity position with tentatively neutral or even positive effects (HQLA improve liquidity management). Capital × Leverage −Nonrisky assets (bonds) drive leverage and further improve capital. The double effect from leverage and capital on loans may be negative. Long term liquidity × Short term liquidity −Long term liquidity and short term liquidity are two complementary requirements for liquidity. Their joint implementation may impose an additional constraint. Long term liquidity × Leverage −Banks achieve both long term liquidity and leverage based on liquid nonrisky assets. Loans do not comply with this; the joint effect may be negative. Short term liquidity × Leverage ± Short term liquidity includes mainly HQLA with a neutral or positive effect on loans, while leverage builds on bonds. The combined effect is mixed. Supervisory & Monetary Capital × Monetary expansion + Well capitalised banks with central bank injections have a double incentive to supply loans. Long term liquidity × Monetary expansion + Banks whose long term funding pressure is alleviated from a central banks injection may tend to increase their loans. Short term liquidity × Monetary expansion + Banks with high liquit assets and additional liquidity from the central bank may tentatively provide more loans. Leverage × Monetary expansion ±Leverage obtained from nonrisky assets and additional money from the central bank have opposite effects on the supply of loans. Supervisory & Rescue Capital × Rescue + For rescued banks, capital will motivate loan supply compared to nonrescued banks. Long term liquidity × Rescue + For rescued banks, long term funding will have a bigger impact on lending compared to nonrescued banks. Short term liquidity × Rescue + For rescued banks, short term liquidity will have a higher impact on lending compared to nonrescued banks. Leverage × Rescue + For rescued banks, leverage will stimulate more lending compared to nonrescued banks. Note: The table discusses effects on the supply of loans from the joint consideration of two policy variables (interaction effects, additional effects). We derive the interaction effects of two variables 1 and 2 based on the question: What marginal impact on the supply (volume) of loans results from both a high level of variable 1 and a high level of variable 2?
3514 International Journal of Finance & Economics, 2025 provide empirical evidence showing when liquidity requirements are added to capital requirements, the benefits of low capital requirements will be weakened as they prompt significant reductions in lending. However, in the Covas and Driscoll(2015) equilibrium framework, imposing both liquidity and capital requirements leads to a sizable contraction in credit since banks become more sensitive to loan price and intend to hold more lowerrisk liquid assets. The authors also mention that the impact of liquidity requirements is much larger than that of capital requirements. Their results suggest that bank capital exerts a significantly positive effect on lending only after large banks have already built up sufficient liquid assets. The interaction effect between short term liquidity ratio and long term liquidity ratio is notable ( 𝛿(SLI ×LLI) = −0.0017). Banks complying with both high SLI and LLI tend to provide less loans than on average. As the single effect from SLI is positive and from LLI is negative, we relate the joint negative impact on loans to the dominance of LLI. Maintaining high ratios of LLI as a balance between long term funding and long term investing is more difficult when banks provide loans, which usually have a midterm and long term duration. Otherwise, SLI mainly build on HQLA, which as a means of liquidity reserve tentatively support the loan business. However, this effect is less expressed than the constraints from LLI. The joint result from long term liquidity requirements and monetary policy is positive ( 𝛾(LLI ×PME) = 0.0020). This is of interest as it means that the negative outcome from long term liquidity requirements (single effect) is more than offset when combining it with extended liquidity supply from the central bank. This result suggests that banks, having funding support from the central bank and thus are able to comply with Basel III long term liquidity requirements, are more likely to provide credit to the real economy. In contrast, the joint effect from high leverage ratio and additional money supply ( 𝛾(LEV ×PME)=−0.0022 ) even increases negative incentives for lending. As has notably become evident from the crisis episodes, additional money from bailout programs was mainly used to support stressed banks. The banks used the money to primarily build up own reserves rather than lending it on to the economy. We can connect this result to the finding from Cozzi etal.(2020). They argue that high gearing of banks and thus higher indebtedness supports the transmission of monetary policy. The effects from providing more liquidity and decreasing interest rates are more pronounced where banks display higher levels of gearing (i.e., lower leverage ratio). Highly geared banks are more vulnerable and restricted in their lending policy. Regarding joint effects between supervisory interaction and rescue policy, the significant difference across rescued and nonrescued banks appears in the tradeoff in short term and long term liquidity as well as leverage. For rescued banks, liquidity and leverage will have a negative impact on lending compared to nonrescued banks ( 𝜃(LEV ×RES)=−0.5527 , 𝜃(LLI ×RES)=−0.3098 , 𝜃(SLI ×RES)=−0.2305 ). This finding indicates that rescued banks prioritise the early implementation of Basel III standards over generating additional lending. It is consistent with the arguments presented by Borio and Disyatat (2011) and Brei, Gambacorta, and von Peter(2013), who assert that rescued banks focus on restoring their regulatory capital ratio without engaging in additional lending. Therefore, we relate it to the fact that rescued banks firstly need to restore their own liquidity and leverage before they can lend on the additional funding to their customers. 4.4 | Robustness Check To ensure the reliability of our findings, we employ a series of robustness tests. Firstly, we examine the estimation results across different estimation periods, taking into account the scale of banking crises. This analysis allows us to evaluate the consistency of our findings under varying crisis scenarios. Secondly, we conduct tests considering different periods that correspond to the presence or absence of the COVID- 19 pandemic. By examining the impact of this global health crisis on our results, we gain insights into the exogenous shock on the observed relationships. Thirdly, we assess the generalizability of our findings by considering both marketbased economies and bankbased economies. This analysis helps us to determine whether the relationships hold consistently across different economic contexts. Fourthly, we verify the results by conducting analyses on representative country groups based on the number of banks. This approach allows us to ascertain whether the observed patterns hold consistently across different groups of countries, accounting for potential variations in the banking sector. Lastly, we review our findings by employing alternative measurements for Basel III ratios. This enables us to assess the robustness of our TABLE 9 | Results of policy interactions: COVID- 19 impact. (2) (3) Panel- FE S- GMM (Eq: 3) (Eq: 3) TIEijt−1×LEVijt−1 −0.0298*** −0.0235*** TIEijt−1×LLIijt−1 0.0252*** 0.0258*** LEVijt−1×SLIijt−1 −0.0025** −0.0015 SLIijt − 1×LLIijt − 1 −0.0009 0.0002 TIEijt−1×PMEijt−1 −0.0042 0.0015 LEVijt−1×PMEijt−1 −0.0044*** −0.0039** LLIijt − 1×PMEijt − 1 0.0035 0.0049* SLIijt−1×PMEijt−1 −0.0006 −0.0009 TIEijt−1×COVijt−1 −0.0913 −0.3224 LEVijt − 1×COVijt − 1 0.0370 0.0878* LLIijt−1×COVijt−1 −0.0582 −0.0654 SLIijt−1×COVijt−1 0.0151 0.0225 Note: The sample period for column 2–5 goes from 1995 to 2006, and the sample period for column 6–9 goes from 2007 to 2021. Panel- FE refers to the panel fixed effects estimator, and S- GMM refers to the system GMM estimator. *Significance at the 10% level. **Significance at the 5% level. ***Significance at the 1% level.
3515 results when using different metrics to assess bank's compliance with Basel III requirements. By conducting these comprehensive tests, we aim to ensure the reliability and validity of our research findings. 4.4.1 | Bank Failures The period spanning from 1995 to 2006 was characterised by local bank failures, whereas the period from 2007 to 2021 witnessed systemic meltdowns, extensive bailouts, and global interventions by central banks. The global financial crisis, along with subsequent regional recessions, posed significant challenges for monetary policy. Throughout both subsample periods, our analysis consistently reveals similar individual effects. However, when examining supervisory interactions, we find that rescued banks exhibit different lending behaviours compared to nonrescued banks. Specifically, the presence of both capital and long term funding in rescued banks encourages lending, whereas the levels of leverage and short term funding in these rescued banks tend to constrain lending but only prior to 2007. Such patterns are not observed in the post- 2007 years with the presence of systemic bank runs (columns 4 and 5 of Table8). Indeed, our findings suggest that the different lending behaviours between rescued banks and nonrescued banks may be overridden by global financial instability. 4.4.2 | COVID- 19 Impact Given the profound impact of the COVID- 19 pandemic, it becomes crucial to investigate its influence on bank lending in G20 countries. While our findings predominantly align with main individual effects and supervisory interactions, it is important to highlight that there is no noticeable difference in the impact of Basel supervision on lending between the period affected by COVID- 19 and non- COVID years (see Table9). During the COVID- 19 period, governments and central banks worldwide implemented a range of monetary and fiscal policies in response to the pandemic. These interventions aimed to mitigate the economic fallout and stabilise financial systems. It is plausible that such policy measures influenced lending practices and TABLE 10 | Results of policy interactions: Economy structure. Bankbased Marketbased (2) (3) (4) (5) (6) (7) (8) (9) Panel- FE S- GMM Panel- FE S- GMM Panel- FE S- GMM Panel- FE S- GMM (Eq: 2) (Eq: 2) (Eq: 3) (Eq: 3) (Eq: 2) (Eq: 2) (Eq: 3) (Eq: 3) TIEijt−1×LEVijt−1 −0.0403** −0.0520** −0.0331** −0.0418** −0.0473*** −0.0371** −0.0430*** −0.0111 TIEijt−1×LLIijt−1 0.0323** 0.0481** 0.0362** 0.0556*** −0.0105 −0.0149 −0.0053 −0.0167 TIEijt − 1×SLIijt − 1 0.0246*0.0109 0.0238*0.0133 −0.0048 −0.0048 0.0018 −0.0017 LEVijt−1×LLIijt−1 −0.0137*** −0.0207*** −0.0111*** −0.0177*** −0.0052 −0.0015 −0.0042 0.0015 LEVijt−1×SLIijt−1 0.0037 0.0060*0.0016 0.0036 −0.0010 0.0031 0.0004 0.0026 SLIijt − 1×LLIijt − 1 −0.0029 −0.0055 −0.0021 −0.0033 −0.0033 −0.0034 −0.0043*−0.0032 TIEijt−1×PMEijt−1 0.1090 0.0359 −0.0168 −0.0107 LEVijt−1×PMEijt−1 0.0120 0.0038 −0.0040 −0.0022 LLIijt−1×PMEijt−1 −0.0036 −0.0576 0.0002 0.0052 SLIijt−1×PMEijt−1 0.0054 −0.0099 −0.0049 −0.0018 TIEijt − 1×RESijt − 1 −5.7372 −8.2553*7.8793*** 4.4122 LEVijt−1×RESijt−1 0.5778 1.4454*−0.0093 0.0473 LLIijt−1×RESijt−1 −1.0437 −2.0795** −0.9230 −0.6267 SLIijt−1×RESijt−1 2.9015** 3.7177** 0.0008 0.0258 Note: The sample period for column 2–5 goes from 1995 to 2006, and the sample period for column 6–9 goes from 2007 to 2021. Panel- FE refers to the panel fixed effects estimator, and S- GMM refers to the system GMM estimator. *Significance at the 10% level. **Significance at the 5% level. ***Significance at the 1% level.
3516 International Journal of Finance & Economics, 2025 mitigated the differential impact of Basel supervision during the COVID- 19 period. 4.4.3 | Economic Structure Among the G20 countries, certain nations like Germany and France have bankbased economies characterised by the significant role of banks in the financial system and corporate financing. In contrast, countries like the United Kingdom and the United States are recognised as marketbased economies, with money markets and capital markets playing a central role. Consequently, conducting a comparative analysis of policy interactions on bank lending between these distinct economic structures can offer valuable insights. Table10 indicates that policy interactions hold greater significance in bankbased economies. This suggests that in such economies, the relationship and interaction between policy measures and bank lending are more noticeable. Conversely, in marketbased economies, the impact of policy interventions on bank lending may be relatively diminished when compared to other influential factors like market dynamics. 4.4.4 | Representative Countries Given that more than 50% of the banks in our sample are from the United States and over 12% are Japanese banks, it is crucial to examine whether the regulatory impact on bank lending is consistent between US (Japan) and the non- US (non- Japan) countries. Our analysis reveals interesting findings in this regard. Specifically, it demonstrates that both the individual and combined effects on bank lending are more significant for US banks compared to non- US banks. Similarly, the policy impact on non- Japanese banks is more notable than on Japanese banks (see Table11). These observations suggest that the regulatory influence on bank lending may vary between the US (Japan) and non- US (non- Japan) countries. 4.4.5 | Alternative Bank Ratios To address concerns related to the arbitrary selection of capital ratio and leverage ratio, our analysis incorporates alternative measures. We substitute regulatory capital with total capital and utilise the ratio of total capital to total assets as a replacement TABLE 11 | Results of policy interactions: Representative countries. US banks Non- US banks Japanese banks Non- Japanese banks (2) (3) (4) (5) (6) (7) (8) (9) Panel- FE S- GMM Panel- FE S- GMM Panel- FE S- GMM Panel- FE S- GMM (Eq: 3) (Eq: 3) (Eq: 3) (Eq: 3) (Eq: 3) (Eq: 3) (Eq: 3) (Eq: 3) TIEijt−1×LEVijt−1 −0.0480*** −0.0298*** −0.0106*−0.0198** −0.0239 −0.0250 −0.0312*** −0.0244*** TIEijt−1×LLIijt−1 0.0086*0.0068 0.0025 0.0044 0.0079 0.0117 0.0027 0.0050 TIEijt − 1×SLIijt − 1 0.0421*** 0.0309*** 0.0166*0.0287*** −0.0127 0.0163 0.0271*** 0.0216*** LEVijt−1×LLIijt−1 −0.0040** −0.0016 −0.0016 −0.0016 −0.0012 −0.0025 −0.0023** −0.0009 LEVijt−1×SLIijt−1 0.0049** 0.0049 −0.0023 0.0004 −0.0020 0.0013 −0.0001 −0.0001 SLIijt − 1×LLIijt − 1 −0.0027 −0.0008 −0.0023 −0.0048*0.0007 −0.0043 −0.0035** −0.0011 TIEijt−1×PMEijt−1 −0.0435*−0.0239 −0.0059 −0.0024 −0.0002 0.0121 0.0021 0.0091 LEVijt−1×PMEijt−1 −0.0099 −0.0036 −0.0032** −0.0037** −0.0277*** −0.0252*** 0.0002 0.0001 LLIijt−1×PMEijt−1 0.0060 0.0043 0.0020 0.0045*0.0038 0.0058 0.0030 0.0021 SLIijt−1×PMEijt−1 −0.0538*−0.0338 0.0027 0.0010 0.0039 0.0023 0.0010 −0.0009 TIEijt − 1×RESijt − 1 −5.1652*** 0.0000 0.9404 −0.6818 1.4748*** 0.0000 0.0022 −0.6428 LEVijt−1×RESijt−1 11.1940*** 1.0352*** −0.1833 0.0175 0.0001 0.1061 −0.0804 0.1109 LLIijt−1×RESijt−1 −6.1159*** −0.7356*** 0.0131 −0.0482 0.0001 0.0001 0.0162 −0.1764 SLIijt−1×RESijt−1 0.0001 −2.5872*** −0.0067 0.0167 0.0001 0.0001 0.0114 0.0172 Note: The sample period goes from 1995 to 2021. Panel- FE refers to the panel fixed effects estimator, and S- GMM refers to the system GMM estimator. *Significance at the 10% level. **Significance at the 5% level. ***Significance at the 1% level. R2 reflects the overall goodness of fit of the data (only Panel- FE). Hansen test (pvalue) assess the identification of the equation (only S- GMM). AR(2) pvalue exams the preseence of second order autocorrelation in residuals (only S- GMM).
3517 for the traditional leverage ratio. The overall regression results, shown in Table12, affirm the robustness of our findings in comparison to the effects obtained from regressions utilising the alternative capital and leverage ratios. This indicates that the chosen alternative measures provide reliable and consistent outcomes that align with the observed effects. 5 | Conclusion This paper studies the impact of multiple policy variables on bank lending. We obtain the total effects as the aggregation of single effects from supervisory, monetary and rescue policies and the interaction (marginal) effects from the parallel application of these policies. To identify the overall effects on lending from two or more jointly applied regulatory variables, we consistently define single effects and combine them into interaction terms. In addition, our analysis also extends to new multipletype interactions of supervisory, monetary and rescue actions. By combining Bankscope and Fitch data, this analysis includes 1052 active banks of the Group of Twenty (G20) countries from 1995 to 2021. We present our key findings as follows. Regarding the single effects of regulatory requirements, capital and short term liquidity requirements provide positive effects for lending, while higher long term liquidity and leverage ratios are associated with negative effects. Notably, the leverage ratio has the strongest single (negative) coefficient among the four considered Basel regulatory requirements. In terms of interaction effects, we find that all supervisory interaction effects based on LEV remain negative. We conclude that the leverage ratio is dominating all other effects. Consequently, supervisors must exercise caution when implementing policy actions related to the leverage ratio, considering the potential ‘overproportional’ impact from LEV . Within the joint effects based on PME , we find that 𝜆(LLI ×PME) has an overall positive impact. This important result suggests that extensive funding from central banks more than offsets the constraints from higher liquidity requirements and results in an overall positive effect on lending. Furthermore, we observe that short term liquidity requirements have a positive effect on bank lending. However, when SLI combined with other requirements, the overall impact turns negative. In addition to our primary analysis, we conducted several alternative tests to ensure the robustness of our research findings. These tests further support the validity and reliability of our conclusions. The findings of our research have important implications for policymakers. Firstly, it is recommended to promote capital and short term liquidity requirements as they contribute to strengthening banks and facilitating lending activities. However, careful consideration should be given to the implementation of long term liquidity and leverage requirements, as they may have adverse effects on lending. Policymakers should conduct thorough evaluations to assess the potential impact of the leverage ratio on lending and ensure that it does not disproportionately hinder banks' lending activities. The positive effects of long term liquidity requirements with monetary policy as well as negative effects of the policy combination with short term liqudity requirements underscore the need for caution when multiple regulatory policies are applied simultaneously, as their interaction can lead to contrasting effects on bank lending. In light of these findings, we argue that policymakers should prioritise a comprehensive understanding of the dynamics of policy interactions when making decisions. This understanding is crucial for formulating optimal policies that achieve a balance between promoting lending and maintaining financial stability. By considering the TABLE 12 | Results of policy interactions: Alternative Basel III ratios. (2) (3) (4) (5) Panel- FE S- GMM Panel- FE S- GMM (Eq: 2) (Eq: 2) (Eq: 3) (Eq: 3) TIEijt−1×LEVijt−1 −0.7632 −17.7662 −0.9961 −14.3150 TIEijt−1×LLIijt−1 −0.0105*** −0.0121*** −0.0088*** −0.0076** TIEijt−1×LLIijt−1 0.0104** 0.0135** 0.0145*** 0.0171*** LEVijt − 1×LLIijt − 1 21.9011*** 19.7059*** 18.5934*** 16.5930*** LEVijt−1×SLIijt−1 21.6366 −223.1656*** 22.5978 −176.1968** SLIijt−1×LLIijt−1 −0.0016 −0.0004 −0.0013 0.0001 TIEijt − 1×PMEijt − 1 0.0020 −0.0003 LEVijt−1×PMEijt−1 0.1885 0.4776 LLIijt−1×PMEijt−1 0.0008 0.0006 SLIijt−1×PMEijt−1 −0.0013 −0.0068*** Note: The sample period goes from 1995 to 2021. Panel- FE refers to the panel fixed effects estimator, and S- GMM refers to the system GMM estimator. **Significance at the 5% level. ***Significance at the 1% level.
3518 International Journal of Finance & Economics, 2025 specific effects and potential interactions of various regulatory requirements, policymakers can ensure a wellinformed and effective approach to policy implementation. Acknowledgements The authors are grateful to the editor, Professor Jia Liu, and two anonymous reviewers for their constructive comments and suggestions. Their support has greatly improved this paper. Conflicts of Interest The authors declare no conflicts of interest. Data Availability Statement The data that support the findings will be available in Bureau van Dijk and Fitch Solution following an embargo from the date of publication to allow for commercialization of research findings. Endnotes 1 Similarly, Cecchetti(2015) distinguishes capital regulation, liquidity regulation and resolution mechanisms. 2 A range of additional requirements has been incorporated to make the business of banks more transparent and compliant with the objectives of stakeholders. Among others, pillars 2 and 3, representing the supervisory review and market discipline approach of the Basel framework, aim to enhance the efficiency and transparency of banks´ management and organisation (BCBS2004). The Markets in Financial Instruments Directive (MiFID) and the Payment Service Directive (PSD) of the European Union (EU) require measures to ensure that banks services are aligned to the need of their customers (Busch 2017; Yeoh 2019). The United States and other countries have adopted similar measures as for example the Dodd- Frank Wall Street Reform and the Consumer Protection Act (Baker, Cumming, and Jagtiani2017). However, as these requirements focus on organisational and reporting obligations and do not directly affect the capital and liquidity or the lending business of banks, they are not considered in this study. 3 We compare the bank names and financial statement data in 2014 from both Bankscope and FitchConnect. They are quite similar, so we complement our Bankscope data with Fitch data from 2014 afterwards. 4 In this study, we do not consider corporate bonds as part of the lending business of banks as it is not obvious if the banks purchased these bonds at the time of emission or later from the secondary market. In addition, loans are considered as the genuine and relationshipbased lending from banks. 5 We also checked the availability of further capital ratios from Bankscope and Fitch, including Fitch Core Capital, Fitch Eligible Capital, Tangible Common Equity, Tier 1 Regulatory Capital, Core Tier 1 Regulatory Capital Ratio. 6 BCBS(2017) and BCBS(2011) introduced the liquidity coverage ratio LCR and net stable funding ratio NSFR. The LCR requires banks to have sufficient highquality liquid assets to withstand a 30- day stressed short term funding scenario. The NSFR evaluates banks' long term liquidity adequacy requiring the available amount of stable funding (ASF) to be larger than the required amount of stable funding (RSF). It takes into account the entire balance sheet and provides incentives for banks to use stable sources of funding. 7 Illiquid assets = Total assets- Liquid assets. 8 We also exam the data on deposit insurance collected from the World Bank (Demirguc- Kunt, Kane, and Laeven2014). The dataset includes the date of inception of deposit insurance by country. Hence, we define the deposit insurance dummy ( DIP ) to be 1 if the observations correspond to the year equal to or beyond the year of inception. However, we dropped the deposit insurance dummy because most of the countries have already deposit insurance schemes in place (mean value of DIP is 0.95). 9 Im- Pesaran- Shin test for cross sectional variables, and Dickey- Fuller test for the time series. 10 We calculate those coefficients that are at least at the 5% significance level. References Allen, F., E. Carletti, I. Goldstein, and A. Leonello. 2015. “Moral Hazard and Government Guarantees in the Banking Industry.” Journal of Financial Regulation 1, no. 1: 30–50. Altunbas, Y., L. Gambacorta, and D. Marques- Ibanez. 2014. “Does Monetary Policy Affect Bank Risk?” International Journal of Central Banking 10, no. 1: 95–136. Arellano, M., and O. Bover. 1995. “Another Look at the Instrumental Variable Estimation of Error- Components Models.” Journal of Econometrics 68, no. 1: 29–51. Ashcraft, A. 2006. “New Evidence on the Lending Channel.” Journal of Money, Credit and Banking 38, no. 3: 751–775. Avalos, F., and E. Mamatzakis. 2023. “Is Bank Resilience Affected by Unconventional Monetary Policy in the Euro Area?” Journal of International Money and Finance 130: 102762. Bagehot, W. 1873. Lombard Street: A Description of the Money Market. Vol. 1962. Homewood, IL: Irwin. Baker, C. M., C. Cumming, and J. A. Jagtiani. 2017. “The Impacts of Financial Regulations: Solvency and Liquidity in the Post- Crisis Period.” Journal of Financial Regulation and Compliance 25, no. 3: 253–270. BCBS. 2004. International Convergence of Capital Measurement and Capital Standards. A Revised Framework. Technical Report. Bank for International Settlements, Basel, Switzerland. BCBS. 2011. “Basel III: A Global Regulatory Framework for More Resilient Banks and Banking Systems.” Technical Report, Bank for International Settlements, Basel, Switzerland. BCBS. 2016. “Literature Review on Integration of Regulatory Capital and Liquidity Instruments.” Working Paper 30, Bank for International Settlements, Basel, Switzerland. BCBS. 2017. “Basel III: Finalising Post- Crisis Reforms.” Technical Report, Bank for International Settlements, Basel, Switzerland. BCBS. 2022. “Evaluation of the Impact and Efficacy of the Basel III Reforms." Technical Report, Basel Committee on Banking Supervision, Bank for International Settlements publication, Basel, Switzerland. Bernanke, B., and C. S. Lown. 1991. “The Credit Crunch.” Brookings Papers on Economic Activity 22, no. 2: 205–248. Berrospide, J. M. 2013. “Bank Liquidity Hoarding and the Financial Crisis: An Empirical Evaluation.” Finance and Economics Discussion Series 2013–03, Board of Governors of the Federal Reserve System (US) 2013: 1–41. Berrospide, J. M., and R. M. Edge. 2010. “The Effects of Bank Capital on Lending: What Do We Know, and What Does It Mean?” International Journal of Central Banking 6, no. 34: 1–50. Bhattacharya, S., and K. Nyborg. 2011. “Bank Bailout Menus.” Working Paper 668, London School of Economics. Birn, M., M. Dietsch, and D. Durant. 2017. “How to Reach all Basel Requirements at the Same Time?” Research Report 28, Banque de France, Paris, France.
3519 Black, K., D. Hancock, and W. Passmore. 2010. “The Bank Lending Channel of Monetary Policy and Its Effect on Mortgage Lending.” FEDS Working Paper 2010-39, Federal Reserve Board of Governors. Blundell, R., and S. Bond. 1998. “Initial Conditions and Moment Restrictions in Dynamic Panel Data Models.” Journal of Econometrics 87, no. 1: 115–143. Borio, C. E. V., and P. Disyatat. 2011. “Global Imbalances and the Financial Crisis: Link or No Link?” Working Paper 346, Bank for International Settlements, Basel, Switerland. Brei, M., L. Gambacorta, and G. von Peter. 2013. “Rescue Packages and Bank Lending.” Journal of Banking and Finance 37: 490–505. Buch, C. M., M. Buchholz, K. Knoll, and B. Weigert. 2021. “Why Macroprudential Policy Matters in a Monetary Union.” Oxford Economic Papers 73, no. 4: 1604–1633. Buch, C. M., and E. Prieto. 2014. “Do Better Capitalized Banks Lend Less? Long- Run Panel Evidence From Germany.” International Finance 17, no. 1: 1–23. Bundesbank, D. 2016. “Bank Recovery and Resolution- The New TLAC and MREL Minimum Requirements.” Deutsche Bundesbank Monthly Report’, no. 68: 63–80. Frankfurt: Deutsche Bundesbank Eurosystem. Busch, D. 2017. “MiFID II and MiFIR: Stricter Rules for the EU Financial Markets.” Law and Financial Markets Review 11, no. 2–3: 126–142. Bussiere, M., J. Cao, J. de Haan, etal. 2021. “The Interaction Between Macroprudential Policy and Monetary Policy: Overview.” Review of International Economics 29, no. 1: 1–19. Cappelletti, G., A. Reghezza, C. R. d'Acri, and M. Spaggiari. 2022. “Compositional Effects of Bank Capital Buffers and Interactions With Monetary Policy.” Journal of Banking and Finance 140: 106530. Carletti, E., I. Goldstein, and A. Leonello. 2019. “The Interdependence of Bank Capital and Liquidity.” BAFFI CAREFIN Working Papers 19128, Centre for Applied Research on International Markets, Banking, Finance and Regulation, Italy. Cecchetti, S. G. 2015. “The Road to Financial Stability: Capital Regulation, Liquidity Regulation, and Resolution.” International Journal of Central Banking 11, no. 3: 127–139. Claessens, S., K. Habermeier, E. Nier, H. Kang, T. Mancini- Griffoli, and F. Valencia. 2013. “The Interaction of Monetary and Macroprudential Policies.” Technical Report, International Monetary Fund, IMF Policy Papers. Cornett, M. M., J. J. McNutt, P. E. Strahan, and H. Tehranian. 2011. “Liquidity Risk Management and Credit Supply in the Financial Crisis.” Journal of Financial Economics 101, no. 2: 297–312. Cortes, G. S., G. P. Gao, F. B. Silva, and Z. Song. 2022. “Unconventional Monetary Policy and Disaster Risk: Evidence From the Subprime and Covid- 19 Crises.” Journal of International Money and Finance 122: 102543. Covas, F., and J. C. Driscoll. 2015. “Bank Liquidity and Capital Regulation in General Equilibrium.” Working Paper no. 2014-85, Washington, DC: Division of Monetary Affairs, Federal Reserve Board. Cozzi, G., M. Darracq Paries, P. Karadi, etal. 2020. “Macroprudential Policy Measures: Macroeconomic Impact and Interaction With Monetary Policy.” Working Paper Series 2376, European Central Bank. De Nicolo, G., A. Gamba, and M. Lucchetta. 2014. “Microprudential Regulation in a Dynamic Model of Banking.” Review of Financial Studies 27, no. 7: 2097–2138. Demirguc- Kunt, A., E. Kane, and L. Laeven. 2014. “Deposit Insurance Database Policy Research.” Working Paper WPS 6934, World Bank Group, Washington, DC. DeYoung, R., I. Distinguin, and A. Tarazi. 2018. “The Joint Regulation of Bank Liquidity and Bank Capital.” Journal of Financial Intermediation 34: 32–46. Assessing Banking Regulation During the Obama Era. Diamond, D. W. 2001. “Should Banks Be Recapitalized?” Federal Reserve Bank of Richmond Economic Quarterly 87, no. 4: 71–96. Diamond, D. W., and R. G. Rajan. 2000. “A Theory of Bank Capital.” Journal of Finance 55, no. 6: 2431–2465. Distinguin, I., C. Roulet, and A. Tarazi. 2013. “Bank Regulatory Capital and Liquidity: Evidence From US and European Publicly Traded Banks.” Journal of Banking & Finance 37, no. 9: 3295–3317. htt p s:// w ww. s cien c e di r e ct . c om / s cien c e/ a r tic le/ pi i/ S 0 378 42 6 61 3002100. EBA. 2015. A Report by the EBA Banking Stakeholder Group. London: European Banking Authority Banking Stakeholder Group, Chapter Proportionality in Bank Regulation. ECB. 2022. “Annual Consolidated Balance Sheet of the Eurosystem.” Technical Report, Monetary policy operations, European Central Bank, Frankfurt, Germany. Ehrmann, M., and A. Worms. 2004. “Bank Networks and Monetary Policy Transmission.” Journal of the European Economic Association 2, no. 6: 1148–1171. Evgenidis, A., and S. Papadamou. 2021. “The Impact of Unconventional Monetary Policy in the Euro Area. Structural and Scenario Analysis From a Bayesian VAR.” International Journal of Finance and Economics 26, no. 4: 5684–5703. Feridun, M., and A. Ozun. 2020. “Basel IV Implementation: A Review of the Case of the European Union.” Journal of Capital Markets Studies 4, no. 1: 7–24. Fitch Solutions. 2023. “Fitch Connect.” Delivery Platform Financial Data. Gambacorta, L. 2005. “Inside the Bank Lending Channel.” European Economic Review 49, no. 7: 1737–1759. Gambacorta, L., and D. Marques- Ibanez. 2011. “The Bank Lending Channel: Lessons From the Crisis.” Economic Policy 26, no. 66: 135–182. Gambacorta, L., and P. E. Mistrulli. 2004. “Does Bank Capital Affect Lending Behavior?” Journal of Financial Intermediation 13, no. 4: 436–457. Gern, K.- J., N. Jannsen, S. Kooths, and M. Wolters. 2015. “Quantitative Easing in the Euro Area: Transmission Channels and Risks.” Intereconomics 50, no. 4: 206–212. Giannetti, M., and A. Simonov. 2012. “On the Real Effects of Bank Bailouts: Micro Evidence From Japan.” American Economic Journal: Macroeconomics 5, no. 1: 135–167. Hoerova, M., C. Mendicino, K. Nikolov, G. Schepens, and S. V. Heuvel. 2018. “Benefits and Costs of Liquidity Regulation.” Working Paper Series 2169. Frankfurt: European Central Bank. Hugonnier, J., and E. Morellec. 2017. “Bank Capital, Liquid Reserves, and Insolvency Risk.” Journal of Financial Economics 125, no. 2: 266–285. Jabbour, R. S. 2022. “The Fragmentation of International Banking Regulation.” International Journal of Finance and Economics 27, no. 4: 4451–4471. Jaccard, J., and R. Turrisi. 2003. Interaction Effects in Multiple Regression. 2nd ed. Thousand Oaks: Sage. Kapan, T., and C. Minoiu. 2013. “Balance Sheet Strength and Bank Lending During the Global Financial Crisis.” Discussion Papers 33/2013, Frankfurt: Deutsche Bundesbank.
3520 International Journal of Finance & Economics, 2025 Kashyap, A. K., and J. C. Stein. 1995. “The Impact of Monetary Policy on Bank Balance Sheets.” Carnegie- Rochester Conference Series on Public Policy 42: 151–195. Kashyap, A. K., and J. C. Stein. 2000. “What Do a Million Observations on Banks Say About the Transmission of Monetary Policy?” American Economic Review 90, no. 3: 407–428. Kim, D., and W. Sohn. 2017. “The Effect of Bank Capital on Lending: Does Liquidity Matter?” Journal of Banking and Finance 77: 95–107. Kishan, R. P., and T. P. Opiela. 2000. “Bank Size, Bank Capital, and the Bank Lending Channel.” Journal of Money, Credit and Banking 32, no. 1: 121–141. Kladakis, G., L. Chen, and S. K. Bellos. 2022. “Bank Regulation, Supervision and Liquidity Creation.” Journal of International Money and Finance 124: 102629. https:// www. scien cedir ect. com/ scien ce/ artic le/ pii/ S0261 56062 2000328. Krahnen, J.- P., F. Noth, and U. Schüwer. 2017. “Structural Reforms in Banking: The Role of Trading.” Journal of Financial Regulation 3, no. 1: 66–88. Laeven, L., A. Maddaloni, and C. Mendicino. 2022. “Monetary and Macroprudential Policies: Trade- Offs and Interactions.” Technical Report no. 92, European Central Bank, European Central Bank Research Bulletin. Llewellyn, D. T. 2015. Proportionality in Bank Regulation. Vol. 2, 56–71. Vienna: SUERS. Malovana, S., M. Hodula, Z. Gric, and J. Bajzik. 2023. “Macroprudential Policy in Central Banks: Integrated or Separate? Survey Among Academics and Central Bankers.” Journal of Financial Stability 65: 101107. Martin, A., C. Mendicino, and A. Van der Ghote. 2022. “Systemic Risk and Policy Interventions: Monetary and Macroprudential Policy.” Technical Report No. 97, European Central Bank, European Central Bank Research Bulletin. Martins, L. F., J. Batista, and A. Ferreira- Lopes. 2019. “Unconventional Monetary Policies and Bank Credit in the Eurozone: An Events Study Approach.” International Journal of Finance and Economics 3, no. 24: 1210–1224. Mesonnier, J., and A. Monks. 2015. “Did the EBA Capital Exercise Cause a Credit Crunch in the Euro Area?” International Journal of Central Banking 11, no. 3: 75–117. Nicolo, G. D., A. Gamba, and M. Lucchetta. 2012. “Capital Regulation, Liquidity Requirements and Taxation in a Dynamic Model of Banking.” IMF Working Papers 12/72, International Monetary Fund. Palas, M. J. U., and F. Moreira. 2022. “The Impact of Government Assistance on Banks' Efficiency.” International Journal of Finance and Economics 27, no. 2: 1731–1748. Peek, J., and E. Rosengren. 1995. “Bank Regulation and the Credit Crunch.” Journal of Banking and Finance 19, no. 3–4: 679–692. Peura, S., and J. Keppo. 2006. “Optimal Bank Capital With Costly Recapitalization.” Journal of Business 79, no. 4: 2163–2201. Pierret, D. 2015. “Systemic Risk and the Solvency- Liquidity Nexus of Banks.” International Journal of Central Banking 11, no. 3: 193–227. Schmaltz, C., S. Pokutta, T. Heidorn, and S. Andrae. 2014. “How to Make Regulators and Shareholders Happy Under Basel III.” Journal of Banking and Finance 46: 311–325. Van der Ghote, A. 2021. “Interactions and Coordination Between Monetary and Macroprudential Policies.” American Economic Journal: Macroeconomics 13, no. 1: 1–34. van Dijk, B. 2017. “Bankscope.” Global Database. Vo, Q.- A. 2021. “Interactions of Capital and Liquidity Requirements: A Review of the Literature.” Technical Report No. 916, Bank of England, London: Staff Working Paper. Wooldridge, J. 2021. Introductory Econometrics: A Modern Approach. 7th ed. Andover, Hampshire: Cengage Learning, Inc. Yan, M., M. J. Hall, and P. Turner. 2012. “A Cost- Benefit Analysis of Basel III: Some Evidence From the UK.” International Review of Financial Analysis 25: 73–82. Yeoh, P. 2019. “MiFID II Key Concerns.” Journal of Financial Regulation and Compliance 27, no. 1: 110–123.