The Missing Link in Bank Behavior: Deposit Interest Rate Setting Under a Dual-Benchmark Framework—A Literature Review

Abstract

The efficacy of monetary policy depends on an accurate model of bank behavior, yet the existing literature has a significant blind spot: the central role of deposit interest rate setting. This paper argues that the deposit rate is the primary arena where banks’ strategic and asymmetric responses to policy signals are revealed. Motivated by the unique dual-benchmark system in Indonesia, where a prudential deposit insurance rate actively competes with the central bank’s policy rate, this study addresses a conceptual problem with global relevance, namely, how monetary policy transmission functions when confronted with conflicting policy signals. To investigate this gap, this paper employs a Systematic Literature Review (SLR), combined with bibliometric analysis. By synthesizing findings from 63 articles selected via the PRISMA protocol, this review first maps the intellectual structure of the field, confirming that while themes of monetary policy and bank behavior are mature, the crucial dimension of deposit rate setting, particularly within a dual-benchmark context, remains a ‘missing link’. The primary contribution of this study is, therefore, building a conceptual framework that recenters the deposit interest rate as the fundamental indicator for assessing asymmetric bank behavior and identifying policy distortions. The findings provide a structured foundation for future empirical research and offer critical insights for regulators on the implications for monetary policy transmission and financial system stability.

1. Introduction

The literature on deposit interest rate determination sits at the crossroads of monetary economics, microbanking behavior, and financial stability. Understanding how commercial banks set deposit interest rates is crucial because the success of monetary policy depends on accurately modeling bank behavior. A major gap in the current literature is its neglect of the central role of deposit interest rate setting. The significance of deposit interest rates for effective monetary policy is clear: the policy will only work if commercial banks follow the reference interest rate as an instrument. This is the key area where bank behavior—whether rational, strategic, or asymmetric—becomes evident. Without analyzing deposit interest rates, we lose the ability to gauge how banks really respond to policy signals, a concern that is especially important given consistent empirical evidence of imperfect policy transmission (Gambacorta, 2004; Sahuc & Levieuge, 2020).

The urgency of monetary policy, especially in countries with banking-focused financial systems, stems from the fact that the banking sector is the primary way through which central banks influence the economy. When this process is interrupted—for example, due to slow or inconsistent responses from banks—the central bank’s ability to control inflation and promote economic growth is weakened. In this setting, ineffective policy signals pose risks of macroeconomic instability, making it crucial and urgent to ensure that monetary policy functions effectively.

This challenge becomes even more complex in a unique institutional setting like Indonesia, a country with banking dominance exceeding 70% and a dual-benchmark system. Historical analysis shows that monetary shocks have a delayed but significant impact on aggregate demand, indicating structural sluggishness in the system’s responsiveness. Therefore, clear policy signals are crucial, as they require the banking industry to align its pricing with the central bank’s benchmark to prevent misalignment in macroeconomic adjustments (Siregar & Ward, 2002). A strong policy signal is essential, requiring the banking industry to synchronize its pricing with the central bank’s benchmark. Empirical evidence from the era of Bank Indonesia’s Seven-Day Reverse Repo Rate supports this friction, where the markup on deposit interest rates increased significantly higher than the markup on lending rates, showing that banks absorbed policy changes unevenly to protect their margins.

It is these specific institutional features that distinguish Indonesia’s banking system from those of other countries, both in Southeast Asia and globally. While many countries, such as Malaysia (PIDM), the Philippines (PDIC), and the United States (FDIC), have deposit insurance schemes that serve as passive safety nets with nominal guarantee limits, Indonesia’s framework uniquely incorporates a publicly announced guarantee interest rate (IDIC Rate) as a requirement for obtaining protection. This feature fundamentally shifts the deposit insurance scheme from a passive guarantee to an active alternative price anchor, which can directly compete with central bank monetary policy signals. This difference in design creates a “natural laboratory” for studying policy conflicts and their effects on bank behavior, making Indonesia a highly relevant and essential case study.

This study enhances the traditional monetary transmission theory by including prudential benchmarks in the behavioral analysis of deposit rate-setting. Conventional models focus on the policy rate channel, where central bank signals influence lending and investment, yet they overlook how prudential ceilings impact the initial response of banks’ funding costs. In Indonesia’s dual-benchmark environment, the BI Rate functions as a monetary policy signal, while the IDIC Rate acts as a prudential limit that restricts how deposit rates can be adjusted upward or downward. By explicitly connecting these two channels, this study introduces a new behavioral mechanism where prudential regulation jointly determines the effectiveness of monetary transmission.

Recent research suggests that the relationship between banking behavior and monetary policy is increasingly complex and inconsistent. Overall, previous studies have highlighted asymmetries in monetary transmission driven by structural and institutional factors; however, no study has explicitly examined the influence of two policy benchmarks—monetary interest rates and deposit guarantee rates—as the primary drivers of asymmetric bank behavior. Studies such as those by Zhong et al. (2024), Cafiso and Rivolta (2024), and Fernández (2024) suggest that differences in national financial structures, prudential regulations, and the expansion of the non-bank sector lead to variations in the transmission of monetary policy. While Vuong et al. (2024) and Oyadeyi (2024) suggest that macroeconomic uncertainty and financial system maturity lead to asymmetric responses by banks to monetary policy, Yahyaei et al. (2024) add an international dimension by considering cross-border financing channels. These findings are reinforced by Chun (2024), who demonstrates that the transmission of short-term and long-term interest rates operates separately and requires collaboration between monetary and fiscal policies.

To clarify the conceptual framework, it is important to define key terms. “Bank behavior” refers to the strategic decisions made by banks in response to market signals and regulations, especially in setting prices for core products, such as deposits. “Asymmetric behavior” in this context refers to banks’ uneven responses to policy signals, such as adjusting deposit rates more quickly or thoroughly when interest rates rise compared to when they fall. The “benchmark interest rate” is theoretically the fundamental rate set by authorities to signal the direction of policy. The “anchor interest rate” is a behavioral concept that refers to the benchmark interest rate most influential in shaping market expectations in practice. The phenomenon of “dual-price signaling” occurs when these two benchmarks compete, creating a conflict between abstract monetary signals and concrete prudential signals. The dual-benchmark framework in this study is operationalized as an analytical approach that uses two different yet related benchmarks simultaneously to evaluate bank behavior.

Unlike traditional single-reference models, this approach offers a more balanced evaluation by placing two policy dimensions, monetary and prudential, on a unified evaluative scale. This dual reference helps clarify the observed phenomenon, which weakens conventional transmission mechanisms and leads to a “policy tug-of-war” between monetary and prudential objectives. This conflict lies at the core of the ambiguity that this study seeks to resolve, explaining the asymmetric behavior of banks.

To understand this ambiguity, it is essential to distinguish the economic nature of the two interest rates. Bank Indonesia’s policy interest rate (BI Rate) functions as a theoretical benchmark, signaling the direction of monetary policy, managing inflation expectations, and broadly influencing the cost of funds to achieve economic stability. In contrast, the Deposit Insurance Corporation’s guarantee interest rate (IDIC Rate) serves as an anchor rate. Its purpose is not to steer the economy but to maintain depositor confidence and ensure banks’ compliance with guarantee rules. Its nature is more tangible, directly relevant to risk management and competition at the individual bank level.

The dual-benchmark framework, as shown in Figure 1, conceptualizes how two separate policy signals, the IDIC Rate as the prudential ceiling signal and the BI Rate as the monetary policy signal, cooperate to influence the strategic conduct of commercial banks. These two benchmarks reflect conflicting but related policy forces: the IDIC Rate enforces a prudential limitation intended to protect depositor confidence. At the same time, the BI Rate aims to influence the cost of funds and stimulate credit through monetary transmission. A “tug of war” is created between these policy goals, forcing banks to weigh compliance, profitability, and liquidity when determining interest rates.

Figure 1. Dual-benchmark framework and behavioral transmission mechanism. Source: Authors.

Within this dual-benchmark environment, banks exhibit strategic responses that can be either rational or asymmetric, depending on how they prioritize monetary versus prudential signals. This behavioral adjustment directly affects deposit interest rate formation, thereby influencing the efficiency of monetary transmission. Ultimately, the framework illustrates that financial stability is not solely determined by central bank actions but by the interaction between monetary and prudential regimes—and by how banks, as micro-level agents, interpret and react to these dual signals within an inherently complex and constrained environment.

Therefore, the ‘missing link’ identified by this study is not just the lack of study but the absence of a theoretical connection between these two important discourses. The primary study challenge of this study is the difficulty in modeling deposit interest rate setting, as this is the key area where such asymmetric behavior can be measured. This lack of focus on deposit rates obscures the core factors influencing banks’ pricing decisions in a dual-reference system. This study does not conduct new empirical study. Instead, it offers a structured theoretical foundation, or a “State of the Art,” which is essential before more specific empirical investigations of dual-reference systems can be effectively carried out.

The literature has emphasized that the main factors causing asymmetry in bank behavior are differences in sensitivity to monetary policy, market structure, and information. However, no research has examined how monetary interest rates and prudential interest rates, also known as deposit insurance rates, function together to shape asymmetric bank behavior when setting deposit rates. This conceptual distinction is the main focus of this study. Recent studies have shown that variations in bank behavior stem from various structural and policy sources. However, these studies have been limited to systems based on a single benchmark or factors other than bank deposit rates.

Sekmen (2024) found that lending rates react asymmetrically to monetary policy, with faster increases during tightening and slower decreases during easing. This suggests the existence of rate tensions stemming from banks’ strategic actions. Oyadeyi (2024) asserted that differences in financial development levels create asymmetry in monetary transmission in Nigeria, while Bediako et al. (2024) showed that climate shocks have a disproportionate impact on the stability of the banking system in Sub-Saharan Africa. On the other hand, Kim and Marchesiani (2024) identified information asymmetry between the central bank and market participants as a cause of distortions in money demand, while De la Horra et al. (2024) showed that unconventional monetary policy has asymmetric effects on corporate investment depending on capital rigidities and market power. Li et al. (2024) added a technological dimension, where fintech innovation creates productivity and risk asymmetries through the too-big-to-fail effect.

It is crucial to clarify Indonesia’s role in this study’s context. This paper does not argue that findings from global literature reviews can be directly applied to the Indonesian banking sector. Instead, we use Indonesia’s unique dual-benchmark system as a case study to emphasize important conceptual issues that are relevant worldwide but have not been widely explored. The limited study on Indonesia within the sample highlights the study gap this paper aims to fill. Therefore, our objective is not to generalize global findings to Indonesia but rather to utilize Indonesia’s anomaly as a justification for the need to develop a conceptual framework currently lacking in the global literature.

Despite extensive research on interest rate transmission and credit allocation, most studies view deposit rate-setting as a passive outcome rather than a strategic decision. Existing frameworks rarely incorporate prudential benchmarks, such as the deposit insurance ceiling, which systematically influence banks’ responses to monetary policy. This omission creates a conceptual blind spot that this study addresses by developing a dual-benchmark behavioral model focused on deposit rate-setting, which serves as the missing link in monetary transmission.

Specifically, this study aims to develop a model of bank behavior, examine the dimensions and metrics identified in the literature, and assess how deposit interest rates function as a key indicator of banks’ responses to dual policy signals. By combining these areas, this paper clarifies the often fragmented understanding of bank behavior and emphasizes the importance of deposit interest rate dynamics as the primary analytical focus. Practically, this study offers valuable insights for regulators, policymakers, and financial institutions to develop more effective interest rate policies, enhance risk management frameworks, and anticipate the impact of guaranteed interest rates. Ultimately, these findings aim to promote a more stable, transparent, and sustainable financial system in banking-dominated countries.

Based on the identified gaps, future studies should focus on empirically testing this conceptual framework. Future econometric studies should explicitly model the simultaneous effects of monetary policy benchmarks and deposit insurance benchmarks, especially in developing countries that are underrepresented in the literature, to quantitatively validate the usefulness of deposit interest rate analysis in detecting policy distortions. To guide readers, this article is organized into five sections: following this Introduction, Section 2 explains the methodological approach, Section 3 presents the results and discussion, and Section 4 summarizes the conclusions and outlines future study directions.

2. Methodological Approach

2.1. Methodology

The Systematic Literature Review (SLR) method, combined with bibliometric analysis, was used to identify the latest trends and map the intellectual structure within a specific field of study. This method adheres to the PRISMA guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses), as updated by Page et al. (2021). Specifically, the protocol follows the PRISMA 2020 item checklist (Page et al., 2021). The completed PRISMA 2020 item checklist is provided as Supplementary Materials. It combines citation analysis and co-occurrence methods to identify relationships between themes and keywords. The software used in the analysis process is VOSviewer version 1.6.20 (Van Eck & Waltman, 2010), which enables the visualization of connectivity between studies and the creation of bibliometric networks.

Bibliographic data were collected from the Scopus and Web of Science databases. Then, Echchakoui (2020) organized the data in Excel to simplify coding and classification. This combination of methods allows the study to provide a detailed overview of the knowledge landscape and its connections. It will offer a comprehensive understanding of study progress in this field (Donthu et al., 2021; Mukherjee et al., 2022). This study is registered with the International Platform of Registered Systematic Review and Meta-analysis Protocols (INPLASY), registration number INPLASY202580077.

2.2. Search Equation

A search equation is not just a list of keywords but is strategically crafted to ensure that search results are comprehensive and relevant. The results from the search equation were then processed further in the PRISMA stage, which connected the search phase and the literature sorting phase. Figure 2 shows the design of the search equation and the search databases used, including Scopus and WOS.

Figure 2. Search equation design. Source: Authors.

This systematic review covers studies on banking and behavior, drawing on scientific works from various fields such as business, management, economics, finance, and accounting. To ensure thorough coverage, this study used a two-step approach in its literature search. The Scopus database served as the main source because of its broad scope and high relevance to the study topic. Additionally, the Web of Science (WoS) database was used as a validation step to test the saturation of literature—that is, to confirm that no significant study groups were missed. The results of this validation showed a high overlap between the two databases, with only a few new relevant articles from WoS. This finding is important because the high redundancy confirms that the primary search in Scopus has thoroughly captured the main discussion in this field, thereby strengthening the validity of the final set of articles analyzed.

The study database is selected from Scopus-indexed articles. The distribution of evaluated studies was according to the journals in which they were published between 2014 and 2024. This journal review highlights the evolution of deposit interest rates over time in response to significant financial and macroeconomic events. During the pre-Global Financial Crisis period (2000–2007), deposit pricing behavior was relatively symmetrical, with banks in many developing countries adjusting interest rates in tandem with central bank policy changes, reflecting stable liquidity conditions and limited competitive pressures. The crisis years (2008–2013) introduced structural changes: banks became more conservative, exhibiting stronger downward rigidity in deposit rates as part of a strategy of liquidity hoarding and precautionary funding amid heightened uncertainty.

In Indonesia, the period after 2014 marked a unique shift, with two benchmarks coexisting—the Bank Indonesia Rate (BI Rate) or BI Seven Days Repo Rate (BI7DRR) and the deposit guarantee interest rate (IDIC Rate). Evidence shows that banks respond more strongly to the IDIC Rate, highlighting the impact of regulatory ceilings on asymmetric deposit pricing. Recently, the COVID-19 pandemic (2020–2021) reinforced banks’ reluctance to lower deposit rates despite aggressive monetary easing. In contrast, the post-pandemic tightening cycle (2021–2023) led to faster deposit rate adjustments to secure funding and maintain market share. These temporal dynamics show that deposit interest rate behavior is not static but evolves in response to global crises and domestic changes, emphasizing the importance of considering path asymmetry and dependence in modeling bank behavior.

2.3. Study Selection and Eligibility Criteria

The study selection process was guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework, a methodological protocol designed to ensure transparency, reduce selection bias, and guarantee replicability. As illustrated in the flow diagram (Figure 3), this framework operationalizes the screening process through four distinct stages: (1) identification of initial records from databases; (2) screening based on titles and abstracts to exclude studies that are clearly irrelevant; (3) eligibility assessment through a full-text review of inclusion criteria, namely methods, data, and study focus; and (4) inclusion of final articles into the corpus for analysis. This systematic approach ensures that the final literature collection produced is highly relevant and of high methodological quality for synthesis.

Figure 3. PRISMA refinement procedure. Source: Authors.

To clearly implement these stages, the detailed inclusion and exclusion criteria used during screening and eligibility assessments are outlined in Table 1.

Table 1. Inclusion and exclusion criteria for study selection.

Criteria	Inclusion Criteria (Included)	Exclusion Criteria (Excluded)
Time Range	Articles published between 2014–2024	Articles published before 2014 or after 2024
Document Type	Journal Articles	Books, book chapters
Language	Articles written in English	Articles in languages other than English
Subject Area	Focus on core fields: Economics, Econometrics, and Finance (54%), and Business, Management, and Accounting	Articles from irrelevant fields (e.g., computer science, medicine)
Thematic Focus (Title/Abstract)	Directly discusses “bank behavior”	Studies primarily focused on non-banking (e.g., insurance), pure digital transformation, or general macroeconomics without a direct link to bank behavior
Thematic Focus (Full Text)	Confirmation of alignment with study focus: role of deposit interest rates in bank behavior models	Studies that passed abstract screening but whose full text deviated from the core focus (e.g., only discussing digital transformation)
Accessibility	Full text available and open access	Abstracts only or subscription-locked articles

This procedure ensured that all literature analyzed was relevant to the study topic, met the established inclusion criteria, and had adequate methodological quality. The complete four-step process of study selection is detailed in the PRISMA flow diagram presented in Figure 3.

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) framework was employed to present the results of the literature screening, as illustrated in Figure 3. The PRISMA diagram delineates a structured, multi-stage process comprising identification, screening, eligibility assessment, and inclusion, ensuring that all reviewed studies are relevant to the research topic, meet the predefined inclusion criteria, and possess sufficient methodological rigor. This systematic approach enhances transparency, minimizes selection bias, and facilitates replication in future research. The process begins with the identification stage, during which potentially relevant articles are gathered from several leading academic databases using a broad search strategy designed to prevent the omission of key studies. It proceeds to the screening stage, where titles and abstracts are reviewed to exclude papers outside the study’s scope, such as those focusing on non-banking financial sectors or digital transformation. Subsequently, the eligibility assessment involves an in-depth review of each remaining article to verify that its objectives, data, and methods align with the inclusion criteria. Studies failing to meet methodological standards or deviating from the core topic are excluded. Finally, the inclusion stage produces a refined set of high-quality, thematically relevant articles forming the foundation for the systematic review and subsequent synthesis.

Building on this framework, it is important to clarify the specific scope and objectives that determine our sample size. The aim of this Systematic Literature Review is not to perform a quantitative meta-analysis where a larger sample would be necessary. Instead, our focus is on conducting a qualitative and thematic synthesis to uncover deep conceptual gaps within a well-defined core body of literature. Therefore, the final set of 63 high-quality articles, selected through a rigorous PRISMA protocol, is considered solid and appropriate for our purposes. This sample size is sufficient to map the main intellectual structures in the field while remaining focused enough to highlight the critical neglect of dual frameworks, which is the primary contribution of this study.

The reduction from 9759 records to 63 final studies resulted from a thorough and systematic multi-stage screening process designed to identify the most relevant literature for the study objectives. This process can be summarized as follows: initially, a broad scope was intentionally used to include all potentially relevant literature. However, this number was significantly decreased after applying specific inclusion and exclusion criteria. The initial filter limited the search to the 2014–2024 period to ensure current relevance and focused on core fields like Economics, Finance, and Business Management. Next, technical filters were applied to include only English-language journal articles that are available in full text and open access, ensuring accessibility and peer-review quality.

After an initial automated screening, a more detailed manual review is carried out. Each abstract of the remaining articles is carefully examined to exclude studies that do not directly address bank behavior or interest rate setting. Finally, the full texts of the articles that pass this stage are read thoroughly to confirm their alignment with the specific focus of this study, which is the role of deposit interest rates in bank behavior models, particularly in the context of dual benchmarks. This study uses a unique dual-benchmark system, with Indonesia serving as a motivating case study, to highlight relevant conceptual issues worldwide. It is this meticulous elimination process at each step that ensures the final set of 63 articles is the most focused, relevant, and high-quality collection of literature to answer the study questions.

3. Results and Discussion

This section presents the results obtained from the PRISMA selection process and subsequent bibliometric analysis. We begin by summarizing the descriptive characteristics of the included literature, followed by an in-depth mapping of the study landscape using VOSviewer.

3.1. Descriptive Analysis of Included Studies

The analysis in this chapter starts with a descriptive profile of the chosen literature collection to establish a basic context before moving to a deeper thematic examination. Initially, it is important to understand the landscape of studies in the selected literature. Examining the journals where this study appeared offers key insights into the main disciplines influencing the discourse and identifies the most prominent academic forums on this topic. Table 2 shows the distribution of the 63 articles across their study journals. This will be followed by additional descriptive statistics, such as distribution by year and geographic focus, to give a thorough overview.

Table 2. Studies published by journal during the period 2014–2024.

Journal	Numbers of Studies
European Economic Review	7
Applied Economics	4
Empirical Economics	3
Cogent Economics and Finance	2
Economic Research-Ekonomska Istrazivanja	2
International Journal of Finance and Economics	2
International Review of Financial Analysis	2
Journal of Banking and Finance	2
Journal of Financial Intermediation	2
Acta Oeconomica	1
Asian Academy of Management Journal of Accounting and Finance	1
Economic Modelling	1
Economic Notes	1
Economica	1
Economies	1
European Journal of Finance	1
Gadjah Mada International Journal of Business	1
German Economic Review	1
International Journal of Financial Studies	1
Journal of Behavioral and Experimental Finance	1
Journal of Economic Behavior and Organization	1
Journal of Economic Surveys	1
Journal of Financial Economics	1
Journal of Financial Services Research	1
Journal of Financial Stability	1
Journal of International Economics	1
Journal of Monetary Economics	1
Journal of Money, Credit and Banking	1
Journal of Risk and Financial Management	1
Latin American Journal of Central Banking	1
Margin	1
Mediterranean Journal of Social Sciences	1
Open Economies Review	1
Oxford Economic Papers	1
Philippine Review of Economics	1
Prague Economic Papers	1
Public Finance Quarterly	1
Quarterly Journal of Economics	1
Review of Economics and Political Science	1
Review of Quantitative Finance and Accounting	1
Revista Mexicana de Economia y Finanzas Nueva Epoca	1
Singapore Economic Review	1
Structural Change and Economic Dynamics	1
Studies in Nonlinear Dynamics and Econometrics	1
Bulletin Of Economic Research	1

Source: Authors.

Table 2 presents the contributions of the evaluated study based on academic journals for the period 2014–2024. With seven studies ranked among the top, the European Economic Review affirms its important role in disseminating leading studies in economics and finance. Other journals, such as Empirical Economics (three studies), the International Journal of Finance and Economics, the International Review of Financial Analysis, and the Journal of Banking and Finance (two studies each), also made significant contributions by combining theoretical and practical studies.

Studies in these global and regional journals reflect the multidisciplinary and geographical scope of the latest study in banking, finance, and economics. Although there are journals that each produce one study, they cover a variety of topics, such as macroeconomic theory (Journal of Monetary Economics, Quarterly Journal of Economics), financial stability (Journal of Financial Stability), or regional perspectives (Gadjah Mada International Journal of Business, Philippine Review of Economics, Latin American Journal of Central Banking). Despite its thematic diversity, banking behavior, monetary policy, and interest rate dynamics are the main themes of this distribution pattern.

All journals included in this study are from two subject areas: Economics, Econometrics, and Finance (54%) and Business, Management, and Accounting (34%). In the Business, Management, and Accounting group, the main focus of the study is on bank behavior, especially business practices and deposit interest rate setting strategies. For example, it looks at shifts in bank intermediation models during the low-interest-rate era (Brei et al., 2020), household deposit mobility in response to interest rate differentials (Gerritsen & Bikker, 2018), differences in credit behavior based on ownership and crisis phase (Haldarov et al., 2020), and how regulation affects financing decisions (Jones et al., 2022). In the financial, economic, and econometric sections, the discussion emphasizes monetary policy, interest rate dynamics, deposit insurance, and asymmetric transmission. Topics include the effectiveness of negative interest rates and tiering (Altavilla et al., 2022; Fuster et al., 2024; Demiralp et al., 2021), deposit channels as transmission channels (Drechsler et al., 2017), and meta-analyses of interest rates.

This pattern illustrates how microperspectives on bank behavior and macroperspectives on policy and financial system stability are integrated. After mapping the outlets of studies, the analysis shifts to the temporal dimension to trace the intellectual development of this topic. Annual studies trends not only reflect growing academic interest but also act as a proxy for the demand for new knowledge. An increase in the number of studies in certain years often relates to real-world events from earlier periods, making it a valid indicator of how the study agenda is influenced by the most pressing economic and policy challenges of the time.

Figure 4 illustrates the annual distribution of reviewed studies from 2014 to 2024, showing an increasing trend in the number of studies published in the last decade, reflecting growing academic interest in this topic. During the initial period (2014–2016), the number of studies per year was relatively low, fluctuating between 2 and 4 studies. An increase emerged in 2017 and reached a significant surge in 2018 with seven studies. Although there was a decline in 2019, the trend resumed its upward trajectory, particularly from 2020 to its peak in 2022 and 2023, with each year seeing nine studies. In 2024, the number of studies decreased slightly to seven but remained higher than in the early part of the decade. The surge in studies during the 2020–2023 period coincided with increased global economic uncertainty and regulatory transformations, which likely drove an increase in studies on bank behavior and deposit rate setting. This trend indicates sustained attention to this field of study, underscoring its relevance for continued study from various perspectives, particularly in deposit rate determination under the dual-benchmark framework.

Figure 4. Reviewed studies by year. Source: Authors.

Based on the distribution of countries in the data (see Table 3), most studies were conducted in the cross-country category with 15 studies, followed by the Euro Area with 11 studies and the United States (US) with eight studies. The United Kingdom (UK) came next with four studies, while Brazil had three. Countries such as China, Germany, India, Indonesia, Mexico, the Netherlands, and Nigeria each have two studies. Meanwhile, several other countries, including Colombia, the Czech Republic, France, the Philippines, Singapore, South Africa, Switzerland, Turkey, and Vietnam, each contribute only one study. This pattern suggests that studies in this field tend to focus heavily on cross-country analysis and major economic regions, such as the Euro Area and the US, with relatively limited representation from individual developing countries.

Table 3. Geographical scope of the analyzed studies.

Country	Number of Studies (N)	% Number of Studies
Cross country	15	24%
Euro Area	11	17%
US	8	13%
Turkey	4	6%
Brazil	3	5%
China	2	3%
German	2	3%
India	2	3%
Indonesia	2	3%
Netherlands	2	3%
UK	2	3%
Colombia	1	2%
Czech Republic	1	2%
France	1	2%
Mexico	1	2%
Nigeria	1	2%
Philippines	1	2%
Singapore	1	2%
South Africa	1	2%
Switzerland	1	2%
Vietnam	1	2%
Total	63	100%

Source: Authors.

Analysis of citations provides an overview of the most significant study contributions in this field. Citation analysis provides an overview of the most significant contributions to the field in this area. Studies with the highest number of citations are considered the most influential works because the number of citations indicates the level of academic recognition and dissemination of ideas in the literature. Table 4 below presents a list of studies with the most significant academic influence.

Table 4. Most influential studies.

Description Paper	Total Citation
Drechsler et al. (2017). This study identifies deposit channels as a new mechanism for monetary policy transmission, showing that increases in the Fed funds rate widen deposit spreads and trigger outflows, particularly in concentrated markets, thereby affecting bank lending contraction through market forces in the deposit market.	311
Buch et al. (2014). This study develops an expanded macroeconomic VAR model incorporating condition factors from approximately 1500 US commercial banks, finding that expansionary monetary shocks increase credit supply and raise prospective risk, with varying transmission depending on bank size, capitalization, liquidity, risk, and exposure to property and consumption credit.	60
Altavilla et al. (2022). This study shows that healthy banks pass on negative interest rates to corporate depositors without reducing deposit volumes, and this policy encourages liquid companies to invest more, challenging the view that conventional monetary policy becomes ineffective at the zero lower bound.	48
Aysan et al. (2017). This study found that implementing macroprudential policies in the Turkish banking sector after the 2008 crisis strengthened depositor discipline and encouraged banks to offer higher interest rates, thereby preventing withdrawals and enhancing financial stability through synergies between regulatory oversight and market mechanisms.	45
Ritz and Walther (2015). This study shows that funding uncertainty in the money market during the 2007–2009 financial crisis explains various bank behaviors, including a decline in lending, increased competition in retail deposits, weak pass-through of policy interest rates, and the need for unconventional monetary policy.	39
Akram and Li (2017). This study found that short-term interest rates are the primary determinant of long-term interest rates in the US. At the same time, higher government debt lowers long-term interest rates in the long run but raises them in the short term, which aligns with Keynes’ view of the central role of central banks.	34
Demiralp et al. (2021). This study demonstrates that the negative interest rate policy (NIRP) in the euro area is expansionary, particularly for banks with high dependence on retail deposits and excess liquidity holdings. Its effectiveness increases when combined with central bank liquidity injections, such as asset purchases.	34
Boyle et al. (2015). This study found that the implementation of deposit insurance during banking crises was only partially effective in preventing bank runs, especially in countries without prior explicit schemes, as withdrawal risk remained high and depositors demanded higher interest premiums. In contrast, long-term relationships with banks and the absence of co-insurance could reduce such risks.	30
Brei et al. (2020). This study shows that prolonged periods of low interest rates have prompted large international banks to shift their activities from interest income to fee-based income and trading, with a more substantial shift among low-capital banks, accompanied by adjustments in funding structures toward deposits, a decline in risk-weighted asset ratios, and a reduction in loan loss provisions consistent with indications of evergreening practices.	27

Source: Authors.

The top ten articles were chosen for citation analysis based on the highest number of citations from the 63 articles in this study (Table 4). These studies were then ranked by the total number of citations received, and the ten leading articles were selected for further review. This approach was used to ensure that articles with the most citations, which have a significant influence on the field of study, were included in the review. The selection was based solely on citation counts, without considering other factors such as journal impact or study year.

3.2. Vos Viewer Result

The PRISMA screening process yielded a core literature corpus that served as the basis for the bibliometric analysis. This analysis was then mapped using VOSviewer to identify patterns of relationships between topics, keywords, and fields of study that emerge from the selected literature. VOSviewer represented the state-of-the-art study in the field of banking behavior. VOSviewer visualized the conceptual landscape formed from these selected studies, highlighting dominant topics, interconnections between themes, and potential study gaps that could become the focus of future study. The following bibliometric visualization, therefore, serves to empirically map the conceptual interconnections and thematic density of the study field.

The patterns in the descriptive section align with the results of the VOSviewer bibliometric mapping, showing that major themes such as monetary policy, interest rates, banking, and lending behavior are dominant in the literature network. The number of studies in macroeconomics, monetary policy, and banking behavior journals indicates that studies on these topics have been widespread over the past decade and center on key issues forming the “backbone” of the global study network. The network diagrams in Figure 5 and Figure 6 offer important insights into the complex interactions between the banking industry and bank behavior as determinants of interest rates on deposits in commercial banks. These visuals illustrate the intricate dynamics of the banking sector and its broader economic impacts by graphically highlighting the connections among key subjects such as the banking industry, monetary policy, lending behavior, interest rates, and financial crises.

Figure 5. Map of keyword co-occurrences. Source: VOSviewer.

Figure 6. Items density visualization map—VOSviewer output. Source: VOSviewer.

The network analysis with a systematic literature review is also enriched by providing a structured and data-driven approach to understanding complex banking phenomena. This methodology facilitates a more comprehensive exploration of how commercial banks behave, particularly in interest rate setting as a model, and its impact on the economy. It also guides future study to address identified gaps and enhance policy effectiveness for regulators in the financial sector. This visualization supports the literature review by highlighting thematic clusters that are concentrated and identifying areas that require further investigation, particularly in relation to banks’ tendencies to deviate from central bank benchmarks and the resulting macroeconomic effects. The way banks establish interest rates on deposits is still not a significant issue in the bank network diagram. Contextualizes the literature review within a broader empirical and theoretical framework, reinforcing the importance of further study on this issue.

To go beyond visual interpretation and quantitatively evaluate the structural importance of each theme within the network, a centrality analysis was performed. Table 5 shows key centrality metrics for the most significant keywords, providing a statistical foundation for their influence and role in scientific discussion. This analysis helps identify which themes act as intellectual ‘centers of gravity’ and which serve as ‘bridges’ connecting different study areas. Understanding these structural roles enables objective confirmation of insights from visual mapping and reinforces the argument about existing study gaps.

Table 5. Theme keywords.

Node	Weighted Degree	Closeness Centrality	Betweenness Centrality	Page Ranks
Monetary Policy	458	0.949153	0.091567	0.054491
Lending Behavior	370	0.692308	0.034958	0.021028
Interest Rate	238	0.610390	0.029781	0.013728
Interest Rates	70	0.529032	0.01369	0.007994
Deposit Insurance	50	1.000000	0.002552	0.005993
Macroeconomics	64	0.531532	0.004543	0.005838
Financial Market	130	0.536842	0.006492	0.005172
Financial Crisis	190	0.639344	0.016391	0.004096
Financial System	92	0.513089	0.004116	0.003844
Empirical Analysis	56	0.516807	0.007126	0.003617
Credit Provision	148	0.527473	0.008141	0.003224
Inflation	70	0.576642	0.005744	0.002921
Exchange Rate	54	0.502488	0.002707	0.002714
Central Bank	176	0.600694	0.009586	0.002367
Banking	398	0.662338	0.009606	0.002246
Cost Analysis	50	0.410658	0.001029	0.002048
Competition (Economics)	54	0.516529	0.000687	0.001956
Banks	60	0.533632	0.000379	0.001865
Bank Lending	104	0.537468	0.000517	0.001677
Capital Controls	48	0.465000	0.001451	0.001549

Source: Gephi. Notes: This table presents network centrality measures for the main keywords (nodes) identified in the bibliometric analysis. Weighted Degree measures the total strength of a keyword’s connections, where a higher value signifies its overall prominence in the literature network. Closeness Centrality measures how quickly a keyword can reach all other keywords, with a higher value (closer to 1) indicating it is highly central and can efficiently disseminate information. Betweenness Centrality measures how often a keyword acts as a “bridge” on the shortest path between other keywords, signifying its importance in connecting disparate study clusters. PageRank measures the influence of a keyword based on the quantity and quality of its connections.

Several keywords have the highest Weighted Degree values in the entire network, according to the results of bibliometric mapping analysis using Gephi 0.10.0. These results show that these words are important in the relevant study landscape. High keyword node values indicate the strength and extent of their connections with other nodes in the network. Based on the highest values, the keywords monetary policy (458), banking (398), lending behavior (370), interest rate (238), and financial crisis (190) occupy the top positions. These topics highlight the dominance of studies focused on monetary policy, the banking system, lending behavior, interest rate mechanisms, and financial crisis dynamics.

The presence of “deposit insurance” (50) in the top list, despite its lower value compared to other major topics, is noteworthy because its “closeness centrality” value reaches 1.000, highlighting its role as an efficient connector to all nodes in the network. Methodologically, closeness centrality measures how close a keyword is on average to all other keywords, so topics like deposit insurance and monetary policy have great potential as connectors that facilitate knowledge flow across themes. Additionally, the relatively high betweenness centrality of monetary policy (0.091567) and lending behavior (0.034958) indicates their strategic roles as connecting paths between clusters. From a PageRank perspective, the highest value is also found in monetary policy (0.054491), confirming its status as the most influential topic in the citation and scientific connection network.

The network analysis results indicate that monetary policy, banking, lending practices, and interest rates are vital topics in scholarly discussions. Financial crises and deposit insurance add to the conversation about financial stability, depositor protection, and risk management. This map highlights a strong multidisciplinary focus on how macroeconomic dynamics, banking regulations, and monetary policy transmission mechanisms interact with each other, both in Indonesia and globally.

However, there is a lack of studies on bank behavior, particularly in setting deposit interest rates, especially in the context of two international benchmarks used to evaluate the effectiveness of monetary policy transmitted to bank interest rates. This gap creates opportunities for important scientific contributions, where analyzing the interest rate-setting behavior of banking market leaders can deepen academic understanding and provide a more solid empirical basis for developing effective monetary policy.

3.3. Thematic Study Cluster Analysis

To offer a clearer understanding of the intellectual framework mapped by VOSviewer, 63 articles in the final corpus were manually coded and sorted into four main super-clusters: Monetary Policy, Bank Behavior, Interest Rates, and Deposit Insurance. Table 6 shows the details of these super-clusters in more specific thematic sub-clusters, along with the key references that define each theme. This classification combines the main theoretical models and empirical findings from the literature, describing the primary study areas and how different schools of thought are organized. This table provides an organized overview that complements the network visualization, highlighting specific content within each of the main study domains discussed in this paper.

Table 6. The main theoretical models of monetary policy, bank behavior, interest rate, deposit guarantee, and monetary policy.

Category	Content	References
Monetary Policy	Cluster 1—Deposit Channel	Drechsler et al. (2017); Girotti (2019); Bürgi and Jiang (2023); Gregor et al. (2020)
	Cluster 2—Credit Channel & Bank Lending	Sahin and Berument (2019); Breitenlechner and Nuutilainen (2023); Glindro et al. (2022); Vithessonthi et al. (2017); Michail et al. (2021)
	Cluster 3—Monetary Policy Response	Bui and Kiss (2020); Holm-Hadulla and Thürwächter (2021); Lozano-Espitia and Arias-Rodríguez (2022); Hommes and Makarewicz (2021); Frenkel et al. (2022); Bhaumik et al. (2018); Kwizera (2024)
	Cluster 4—Risk & Financial Stability	Sarkar and Sensarma (2019); Beernaert et al. (2023); Moyo and Phiri (2024); Kurowski and Smaga (2018); Gebauer and Mazelis (2023); Evgenidis and Malliaris (2022)
	Cluster 5—Crisis Conditions	Zhu et al. (2021); Chow and Choy (2023); De Roure et al. (2015)
	Cluster 6—Global Monetary Policy	Aldasoro et al. (2023); Sibande (2024)
Bank Behavior	Cluster 1—Risk, Capital, and Bank Heterogeneity	Buch et al. (2014); Raunig et al. (2016); Boeckx et al. (2020); Jones et al. (2022); Memmel et al. (2016)
	Cluster 2—Competition, Regulation, and Market Structure	Xu et al. (2016); Aydemir (2014); Aysan et al. (2018)
	Cluster 3—Deposits, Transfers, and Customer Responses	Gerritsen and Bikker (2018); Ojeaga and Odejimi (2014); Fuster et al. (2024); Felici et al. (2023)
	Cluster 4—Bank Structure, MSMEs, and Lending Behavior	Aysan et al. (2017)
	Cluster 5—Bank Behavior During Crises and Uncertainty	Ritz and Walther (2015); Raunig et al. (2016); Haldarov et al. (2020)
Interest Rate	Cluster 1—Monetary Policy Transmission and Macroeconomic Effects	Kotb and Proaño (2022); Haubrich et al. (2012); Ha and Quyen (2018); Cotler and Carrillo (2024); Mahrous et al. (2020)
	Cluster 2—Market Structure, Competition, and Risk Perceptions in Interest Rate Setting	Varga (2021); Heinzelmann and Missong (2019); Moyo and Phiri (2024); Da Silva and Baczynski (2024)
	Cluster 3—Low and Negative Interest Rate Environment	Demiralp et al. (2021); Altavilla et al. (2022); Felici et al. (2023); Creel et al. (2016)
	Cluster 4—Banking Behavior, Depositors, and Distributional Impacts	Brei et al. (2020); Beatriz et al. (2022); Cucciniello et al. (2022)
Deposit Insurance	Cluster 1—Depositor Behavior and Risk Preference Shifts	Jameaba (2018); Tran et al. (2024)
	Cluster 2—Deposit Insurance Credibility and Financial Stability	Bonfim and Santos (2023); Boyle et al. (2015)

Source: Authors. Note: This table presents the thematic classification of the 63 articles into four primary super-clusters and their sub-themes, derived from bibliometric mapping and qualitative analysis.

A thematic analysis of the literature, as categorized in Table 6, reveals four main lines of study that form the core of the discourse on bank behavior. The Monetary Policy cluster is the largest, mainly focusing on transmission mechanisms, primarily through the Deposit Channel and the Credit Channel. This area also thoroughly investigates central bank policy responses to various economic conditions and their effects on systemic risk and financial stability. It indicates a mature field of study centered on the macroeconomic effects of central bank actions.

The Banking Behavior Cluster shifts its focus to the microlevel, mainly examining how Risk, Capital, and Bank Heterogeneity affect institutional responses to policy and market changes. Other key sub-themes include the influence of Competition and Regulation on market structure and the dynamics of Deposits and Customer Responses, especially during periods of crisis and uncertainty. Likewise, the Interest Rate cluster looks at monetary transmission from a pricing perspective, emphasizing its macroeconomic impact, the role of market structure in setting interest rates, and the difficulties presented by low or hostile interest rate environments.

Finally, the Deposit Insurance cluster, although smaller, is distinct and highly focused. Study in this area examines how deposit insurance systems influence Depositor Behavior and risk preferences, as well as the critical role of Credibility schemes in Financial Stability. The visual separation of these clusters from the main Monetary Policy discourse confirms the study gap identified in this paper: the direct impact of deposit insurance interest rates as a competing benchmark in bank pricing behavior remains an understudied yet crucial area for investigation.

As a complement to the thematic analysis in Table 6, Figure 7 presents descriptive statistics of the analyzed literature corpus. This bar chart visually maps the distribution of the number of sub-clusters identified within each of the four main super-clusters. This visualization provides a clear quantitative picture of the ‘weight’ and ‘thematic diversity’ of each study domain, graphically illustrating the dominance of the ‘Monetary Policy’ cluster (six sub-clusters) compared to the narrower focus of the ‘Deposit Insurance’ cluster (two sub-clusters).

Figure 7. Descriptive statistics of the analyzed literature corpus. Source: Authors.

3.4. Discussion

3.4.1. The Intellectual Structure of Bank Behavior Research

A Systematic Literature Review (SLR) method, combined with bibliometric analysis, was used to identify recent research trends and map the intellectual structure of the field. The results of the bibliometric analysis using VOSviewer indicate that the four most significant and prominent keywords in the literature network are monetary policy, lending behavior, banks, and interest rates. This cluster structure highlights two dominant domains: (i) macro-level policy design and (ii) micro-level bank responses, which bridge monetary authority and institutional behavior.

Critically, the cluster separations identified in the bibliometric analysis (Figure 5 and Figure 6) are not just visual artifacts. Instead, they directly represent conceptual distinctions found in existing literature. We observe that the “Financial Patterns” cluster (which concentrates on transmission measures, pass-through, and lending behavior) is evolving into a mature research area, yet it remains structurally separate from the “Deposit Insurance” cluster. “Deposit Insurance” appears as an efficient but isolated node, indicating that the literature tends to see it as a passive mechanism for prudential stability rather than a price setter. This statistical distinction offers empirical evidence for our proposed “missing link”: the literature has not incorporated prudential signals (such as the guarantee rate) as a behavioral influence on par with, and possibly competing against, traditional monetary policy signals in shaping bank deposit prices.

The dominance of monetary policy reflects the academic focus on understanding the mechanisms and effectiveness of central bank interest rate policy. Meanwhile, credit allocation behavior reflects academic attention to how banks respond to policy signals and market conditions through credit decisions. Interest rates emerge as a crucial factor directly linking monetary policy, bank behavior, and macroeconomic outcomes. Furthermore, deposit insurance emerged as a key node in a separate cluster, indicating that the study of deposit insurance remains somewhat isolated from the mainstream discourse on monetary policy, despite its strong conceptual links to financial stability and bank decision-making. This isolation implies the absence of an analytical bridge between prudential mechanisms and behavioral aspects of deposit pricing.

The keyword “monetary policy” dominates the core network, having the largest node size and broadest connectivity. It is closely related to interest rates, monetary transmission, pass-through, and transmission mechanisms, which form the intellectual backbone of the field. Bridge nodes, such as lending behavior, central banks, and credit channels, emphasize that the effectiveness of transmission depends on how micro-level decisions internalize macro-level signals.

The dominance of the keywords “monetary policy” and “interest rate” suggests that researchers continue to view monetary policy as the primary driver of bank behavior. This dominance occurs because interest rate policy remains the most direct and measurable channel through which central banks influence funding costs and risk-taking behavior. The limited integration of the topic of deposit insurance in the literature indicates that most previous research treats prudential mechanisms as factors external to monetary policy design, rather than as co-determined behavioral forces.

In the classical framework of monetary transmission, as introduced by Mishkin (1995, 2007), the interest rate channel, particularly through the lending interest rate, is the primary mechanism by which policy actions influence aggregate demand, investment, and output. However, deposit interest rates should be the focus of separate studies, as they are the initial endpoint of monetary transmission; they determine the cost of bank funds, influence the level of liquidity competition, and shape the responsiveness of lending rates to economic policy. Because banks’ cost of funds structures do not readily adapt to changes in monetary policy, monetary transmission is weakened when deposit rates remain rigid or fail to move in line with policy rates. Under these conditions, banks are incentivized to retain third-party funds, preventing immediate reductions in lending rates. This process creates double rigidity in funding and credit distribution. Consequently, policy incentives are delayed in the real sector.

The existing literature also develops in several separate domains. Studies by Bernanke and Gertler (1995) and Mishkin (2007) developed the monetary-transmission and credit-channel framework emphasizing the role of policy rates and credit distribution; Borio (2011) and Allen and Gale (2004) expanded the focus to financial stability and market frictions; Demirgüç-Kunt and Huizinga (2004) and Anginer and Demirguc-Kunt (2018) examined how deposit insurance affects bank-funding behavior and risk-taking incentives; while Sarkar and Sensarma (2019) and Gregor et al. (2020) explored the asymmetry and rigidity in the adjustment of lending and deposit interest rates. However, few studies have successfully integrated all these domains into a unified behavioral framework that links monetary signals (policy rates) with prudential constraints (insured deposit rates).

3.4.2. The Root of Fragmentation: How a Missing Focus on Deposit Rates Complicates Transmission

Given the complexity of monetary transmission, returning to the basic mechanisms is crucial to understanding the overall transmission dynamics. A key element of monetary policy transmission is interest rate dynamics, whose success depends on a complex combination of the macroeconomic framework, market configuration, interest rate anomaly scenarios, and the actions taken by financial institutions and consumers. A comprehensive literature review reveals that the interest rate transmission mechanism and its macroeconomic consequences form the foundation. Changes in policy interest rates significantly affect borrowing costs, which in turn constrain inflation, consumption, investment, and output (Kotb & Proaño, 2022). Interest rates intrinsically govern investment, consumption, and savings behavior by altering borrowing costs, stimulating business projects and consumer spending with lower rates while causing a contraction with higher rates, which in turn influences savings incentives and drives capital reallocation toward sectors like equities and real estate (Bulíř & Vlček, 2015; Dotsis, 2020; Ysmailov, 2021; Lahav & Benzion, 2022).

Due to variations in product categories and loan sizes, these transmission mechanisms do not always follow a linear path (Cotler & Carrillo, 2024). They are often asymmetric depending on existing structural conditions (Mahrous et al., 2020). Interest rate volatility, unpredictable fluctuations over time, introduces uncertainty that complicates investment and financing decisions for economic agents due to unstable borrowing costs, while simultaneously rippling through financial markets to shift the prices of bonds, equities, and other instruments, often prompting central banks to respond by anchoring expectations through forward guidance and policy signaling (Bulíř & Vlček, 2015; Zhang et al., 2020; Mishkin, 2011; Jiang & Fu, 2025).

This complexity is even more pronounced in developing countries, where structural factors play a greater role. Policies in developing countries often address inflation more aggressively and use the exchange rate as an additional tool (Ha & Quyen, 2018). Market structure, risk, and bank competition influence the effectiveness of this transmission. Even in a low-interest-rate environment, risk perception, the level of competition, and market structure can all weaken transmission (Varga, 2021). Risk perception (VDAX) and economic activity (GDP) are crucial for credit pricing (Heinzelmann & Missong, 2019).

The unconventional policy environment and the strategic responses of individual banks further challenge these market conditions. There are specific challenges when interest rates are low and negative, which can squeeze net interest margins and weaken credit transmission because retail deposits cannot fall below zero (Demiralp et al., 2021). However, these policies can still stimulate the economy by rebalancing corporate portfolios (Altavilla et al., 2022). From a policymaker’s perspective, monetary policy behavior is nonlinear, evident in stronger responses to inflation during recessions (Bui & Kiss, 2020), and requires navigating interactions with fiscal policy (Lozano-Espitia & Arias-Rodríguez, 2022), managing public expectations in low-interest regimes (Frenkel et al., 2022), and balancing the inherent conflict between price and financial stability as policies influence default versus market risks (Sarkar & Sensarma, 2019; Kurowski & Smaga, 2018), a complexity further compounded in developing countries where international credit flows transmit global shocks (Aldasoro et al., 2023).

Ultimately, bank and customer behavior are influenced by interest rate changes. Banks with limited capital diversify into non-interest-bearing activities to offset declining margins (Brei et al., 2020), and differences in the treatment of SME customers (Beatriz et al., 2022) add a deeper layer of heterogeneity to the transmission process. These findings collectively suggest that the transmission mechanism is not a consistent mechanical process; it is a highly diverse ecosystem fragmented by market structure, institutional constraints, and strategic responses.

The complex behavior exhibited by financial institutions in response to monetary policy and market conditions arises from an intricate interaction between internal attributes, such as risk, capital, and heterogeneity, where well-capitalized banks show resilience (Boeckx et al., 2020; Raunig et al., 2016) while risk-taking is dynamic (Buch et al., 2014), and external competitive and regulatory structures that shape pricing strategies, evident when rate caps shift competition to innovation (Xu et al., 2016) or market concentration encourages collusion (Aydemir, 2014), a nuance further exemplified by how different business models, like the SME distinctly influence transmission mechanisms (Aysan et al., 2017).

Because of this fragmentation, the analytical focus should be on deposit rate setting, as this is where many different and often conflicting forces converge into a single, visible pricing decision. This decision is a crucial factor in understanding the bank’s actual behavior. Recognizing this fragmentation is a beginning, not an end. It reveals that the fragmentation itself is a symptom of a deeper analytical omission: the behavior of the deposit rate, which serves as the actual ‘missing link’. This omission is not just a gap in empirical research, but a critical failure in the theoretical tools used to connect micro-level banking incentives with macroeconomic policy signals. Identifying this link, therefore, is no longer merely an academic issue but a pressing policy extension. Without an integrated framework centered on this specific variable, policymakers are left to confront a complex system with an incomplete map, unable to grasp the hidden structure of banking decisions fully.

The cluster separation identified in Section 3.4.1 is an empirical sign of this conceptual failure. The visual and statistical gap between the ‘Monetary Policy’ and ‘Deposit Insurance’ clusters confirms the ‘missing link’ that this study focuses on. The literature has not addressed both policy signals within a single, integrated behavioral framework. This failure to integrate these competing signals, specifically how they collide in the arena of deposit rate-setting, is the precise conceptual gap this systematic review seeks to fill.

3.4.3. The Battle for the Benchmark: How Deposit Insurance Creates a Competing Anchor for the Deposit Rate

The traditional role of deposit insurance is primarily to serve as a financial safety net, maintaining financial system stability, especially during times of crisis. It can then shift depositor behavior by shifting depositor preferences from yield-oriented to safety-oriented. Its subsequent role is to prevent bank runs, which are massive withdrawals, by maintaining public confidence in the banking system. A systematic literature review reveals that the establishment of deposit insurance institutions, such as the Deposit Insurance Corporation (IDIC) in Indonesia, significantly shifted depositor behavior from yield-oriented to safety-oriented (Jameaba, 2018). However, this protection does not eliminate the need for caution on the part of the depositor. Even after the crisis, depositors whose funds were insured continued to show sensitivity to the “wake-up call” effect (Tran et al., 2024). The introduction of deposit insurance during the crisis proved only partially effective in preventing large-scale bank runs, so factors such as long-term relationships between banks and customers also play a significant role in mitigating the risk of bank runs (Boyle et al., 2015). The effectiveness of these schemes depends heavily on their credibility. When the credibility of a deposit guarantee scheme is questioned, this can lead to a reallocation of deposits to banks with perceived stronger state guarantees, underscoring the importance of public trust in guarantee institutions (Bonfim & Santos, 2023).

In practice, however, a divergence from the central bank’s policy stance has been observed, reflected in commercial banks’ increasing reliance on alternative reference rates, such as the interest rate set by the Indonesia Deposit Insurance Corporation (IDIC), rather than strictly adhering to the central bank’s benchmark rate. This form of asymmetric banking behavior creates patterns misaligned with the intended goals of monetary policy.

The transformation of Indonesia’s role represents a fundamental difference between the Indonesian system and other systems that only have nominal coverage limits. This has resulted in a shift in the role of IDIC from a passive guarantor to an active alternative price anchor. IDIC interest rates are no longer simply information but rather price signals to which banks must respond. This is the essence of this discussion: the emergence of dual-reference behavior. The “dual-price signaling” phenomenon creates conflicting incentives for banks. Banks face a dilemma between following macroeconomic policy directives and complying with more concrete and operationally relevant prudential rules. Banks rationally prefer signals that are more certain and have direct implications for their operations (i.e., the IDIC Rate). This explains the asymmetric behavior that must be observed.

Because banks are more likely to respond to the IDIC Rate signal than the BI Rate, the dual-benchmark phenomenon weakens the effectiveness of monetary transmission. Due to interest margin pressures, banks increase risk exposure when interest rates rise (Borio & Zhu, 2012). However, the scope for this risk behavior is institutionally limited when a guaranteed rate cap is in place. As a result, policy transmission is not through the risk channel, but through conservative adjustments to deposit rates. The “dual-price signaling” phenomenon creates conflicting incentives for banks. Capital pressures influence bank interest-setting decisions through the capital channel (Van den Heuvel, 2002). Banks with limited capital are more likely to maintain high deposit rates to maintain liquidity than to follow monetary policy. These two mechanisms work together to explain why Indonesia’s asymmetric monetary transmission behavior is caused by the dual-reference system, specifically the BI Rate and the IDIC Rate.

The negative consequences of dual-benchmarking behavior weaken the Central Bank’s signal, as the dominant signal from the Deposit Insurance Corporation (IDIC) can “mute” or weaken the effectiveness of the Central Bank’s signal. This phenomenon creates a “policy tug-of-war” between monetary and financial stability objectives, ultimately disrupting the transmission mechanism of conventional monetary policy. The difference between the central bank’s benchmark interest rate and the interest rate adopted by commercial banks creates a regulatory dilemma. This creates conflicting incentives between the objectives of encouraging investment, consumption, and savings. In response to slow economic growth or instability, the central bank may maintain low interest rates to stimulate investment and consumer spending. At the same time, commercial banks may raise their deposit rates to attract liquidity or align with the benchmark interest rate set by the Indonesian Deposit Insurance Corporation (IDIC), which guarantees the protection of customer funds.

3.4.4. The Missing Link: The Omission of Deposit Rates in Explaining Asymmetric Bank Behavior

Bibliometric mapping using PRISMA and VOSviewer confirms that although the dominant themes revolve around policy rates, pass-through, and the credit channel, the behavior of deposit rates under a dual-benchmark regime remains peripheral. This peripheral position suggests that without modeling deposit rate setting, estimates of the risk of interest rate transmission strength are systematically biased. The asymmetry is evident: responses tend to be faster when the policy rate rises (to preserve deposits) than when it falls (to protect margins). With the existence of a ceiling on the deposit insurance rate, banks may rely more on depositor protection benchmarks than on monetary transmission objectives. The behavior of deposit rates has not been widely treated as a primary study variable for assessing the strength of transmission. Without modeling deposit rate behavior, the estimated strength of interest rate transmission will be biased.

Taken together, the literature consistently highlights the roles of monetary policy, transmission, interest rates, bank behavior, and deposit insurance. However, the dual-benchmark behavior of banks, anchored simultaneously to central bank policy rates and deposit guarantee reference rates, remains a missing link in the discourse. The behavioral dimension of banks operating under a dual-benchmark regime represents a missing link. Explicitly recognizing this blind spot is essential to understanding why monetary policy effectiveness may diverge in contexts such as Indonesia’s dual-benchmark regime.

To strengthen the theoretical foundation, this study can be linked to the literature on hybrid banking models. As explained by Singh et al. (2023), approaches such as the Dominance-Based Rough Set Approach (D-RSA) offer a robust framework for decision-making in situations of incomplete and uncertain information. The core idea of this method is attribute reduction, which simplifies complex systems by identifying and emphasizing the most important or dominant factors. In analogy to a dual-reference system, this approach shows how banks evaluate two policy signals (the BI Rate and the IDIC Rate) that have different levels of dominance in their interest rate-setting process.

This review has clear policy implications: central banks risk overestimating the effectiveness of their policies while prudential authorities act as powerful, competing market signalers. An integrated policy framework is therefore essential to coordinate these signals and prevent a damaging ‘policy tug-of-war’, a conflict where the central bank’s objective to steer the economy is counteracted by banks anchoring their deposit rates to the prudential authority’s benchmark, ultimately weakening monetary transmission.

4. Conclusions

This study aims to develop a coherent conceptual understanding of bank behavior models, specifically by reconceptualizing their role in the context of dual-reference systems. The main objective is to analyze the dimensions and metrics used in the literature to assess bank behavior and, most importantly, to evaluate why deposit interest rates, as a fundamental indicator, have been absent from this framework.

The main finding of this systematic literature review is the confirmation of a theoretical disconnect in the existing literature. The motivating case study in Indonesia, a country with a unique dual-reference system, clearly illustrates the importance of this missing link. Bibliometric analysis shows that study on ‘monetary policy’ and ‘deposit insurance’ runs along two separate academic tracks. It is this structural separation that explains why setting deposit interest rates as the main arena in which asymmetric bank behavior in a dual-reference system can be measured is the “missing link” in current models of bank behavior. By not placing deposit interest rates as the focus of analysis, the fundamental drivers of pricing decisions and asymmetric bank behavior in such systems become invisible. Therefore, this study uses the anomaly in Indonesia not as a single object of study, but as a strong justification for why this global literature review is necessary to construct the conceptual framework that has been missing.

The implications of these findings are significant. Failing to explicitly model deposit rate setting leads to an incomplete understanding of why monetary policy effectiveness might decline. This underscores the need to improve current monetary transmission theories and develop a comprehensive framework that incorporates dual price signals. Practically, these results offer guidance for regulators to establish a coordinated approach where monetary authorities and deposit insurance agencies synchronize their signaling strategies.

This study also highlights the importance of coordinated efforts between the monetary authority (Bank Indonesia) and the prudential authority (IDIC). In a dual-benchmark system, prudential signals, such as the IDIC Rate, may unintentionally overshadow monetary signals, thereby reducing the effectiveness of the BI Rate as a policy anchor. Effective policy coordination requires clear communication and joint calibration between Bank Indonesia and the IDIC, ensuring alignment in timing, magnitude, and goals. More specifically, this collaboration should go beyond mere communication and involve a shared operational mechanism. First, data integration and a real-time risk monitoring dashboard are needed to combine IDIC micro-prudential data with BI macro-prudential data. Second, both institutions should regularly conduct joint modeling and stress testing to assess how changes in the BI Rate and the IDIC Rate impact banking stability and the achievement of monetary policy objectives. Third, joint calibration means establishing a dynamic spread or corridor between the BI Rate and the IDIC Rate, which is periodically adjusted based on a joint systemic risk assessment rather than just individual institutional considerations.

Based on the identified gaps, future studies should focus on empirically testing this conceptual framework. Future econometric studies should explicitly model the simultaneous effects of monetary policy benchmarks and deposit insurance benchmarks, particularly in developing countries that are underrepresented in the literature, to quantitatively validate the usefulness of deposit interest rate analysis in detecting policy distortions.

It is advisable to undertake further, more detailed studies on the effect of bank behavior models by including the process of determining savings interest rates of commercial banks in the evaluation of these models. Such studies should be grounded in scientific data analyzed with econometric techniques to explore the patterns of general bank behavior models in establishing savings interest rates, whether related to current central bank policies or those set by the Deposit Insurance Corporation.

Furthermore, the applicability of this conceptual framework extends beyond conventional commercial banks. The proposed dual-benchmark behavioral model can indeed be applied to other financial institutions, such as Islamic banks (navigating Sharia-compliance profit-sharing rates versus conventional policy rates), microfinance institutions, or cooperative banks, where similar dual constraints or competing prudential benchmarks exist. This potential for cross-application underscores the model’s generality and policy relevance beyond Indonesia’s specific context and suggests a rich avenue for future comparative studies in other dual-rate systems.