1. Introduction
The deposit insurance system (DIS) is one of the fundamental instruments for protecting depositors and preserving the stability of the banking system. Its importance stems from the fact that the stability of banks is largely conditioned by the trust of depositors in the financial system, and that the loss of this trust can result in the destabilisation of the entire banking sector. In this context, the DIS forms an integral component of the broader financial safety net, alongside the central bank’s role as the lender of last resort and the banking regulatory and supervisory framework (Kahn & Santos, 2005).
Given the inherent instability of the banking industry and the vulnerability of the financial system to shocks, it is crucial that countries have adequate crisis prevention and management mechanisms in place (Učkar & Petrović, 2021). In this sense, Ayadi and Lastra (2010) distinguish ex-ante measures, which include regulation, supervision and business transparency, and ex-post measures that are activated in crisis situations. These include the abolition of deposit convertibility, central bank interventions, DIS, insolvency resolution procedures and implicit protection of banks or depositors in cases of systematically important institutions (too big to fail).
The DIS performs two key public functions: preventing bank runs and protecting small depositors (Barth et al., 2006). The latter function is particularly important because small depositors often lack sufficient knowledge and information to properly assess investment risks and the financial health of banks (Faulend & Kraft, 2004).
Accordingly, Carisano (1992) points out that preventing mass withdrawals from banks is the basis for justifying the existence of a DIS. Furthermore, Zolea (2023) emphasised that deposits represent a key source of funding for traditional banking, which further confirms the importance of the insurance system in preserving the stability of the banking sector.
Although DIS has an important stabilising role, its application can cause unwanted effects. Duttagupta and Cashin (2008) warn of the ambiguity of a system that, on the one hand, prevents bank runs, but, on the other, encourages moral hazard. According to Garcia (1996), adverse selection and the principal-agent relationship problem are among the main “traps” of deposit insurance. Moral hazard results from the reduced incentive of depositors to monitor banks and banks’ tendencies to take greater risks (Cullen, 2023; Kundid Novokmet, 2015; Pierret & Howarth, 2023; Protić, 2002). Adverse selection occurs when all banks pay the same premium, regardless of risk, which favours smaller and riskier banks (Garcia, 1999; Huberdeau-Reid & Pennacchi, 2025; Suljić Nikolaj et al., 2022; Wheelock, Kumbhakar, 1995). The problem of principal-agent relationships is further exacerbated if the insurance system operates under political or sectoral influence or if there is a lack of coordination among supervisory institutions (Faulend, 2002). Although numerous studies indicate a higher risk in the behaviour of banks in systems with deposit insurance (Allen et al., 2011; Calomiris & Jaremski, 2016; Dewenter et al., 2018; Gropp & Vesala, 2004; Laeven, 2002; Lambert et al., 2017), its basic function, preventing destabilisation through withdrawal of deposits, remains crucial (Alam et al., 2021; Matutes & Vives, 1996; Martin et al., 2016; Merton, 1977). Therefore, Ayadi and Lastra (2010), and Dávila and Goldstein (2023) point out that a well-designed system, with effective supervision, strengthens confidence in the banking system.
This is confirmed by the Global Financial Crisis (GFC), which, in addition to having a pronounced debt character, also took on the characteristics of a banking crisis in a large number of EU countries (Laeven & Valencia, 2013). It was precisely in such circumstances that it was crucial to ensure the stability of deposits and preserve the confidence of depositors despite the pressures and the distrust generated by the crisis. In response to these challenges, the EU initiated the process of creating a Banking Union, the third pillar of which is a single DIS. The proposal for the EDIS Regulation, which was submitted by the European Commission in 2015 (European Council, 2024), has not yet been formally adopted as a common DIS that would replace national schemes.
The reason for this is that there is still no consensus among the Member States on a common DIS, primarily due to concerns about moral hazard, possible cross-border subsidization and uneven risk distribution (Kuznichenko et al., 2021; Single Resolution Board, 2025). For this reason, the aforementioned topic stimulates the current debate on the role and optimal design of DIS in ensuring financial stability (Arda & Dobler, 2022; Beck et al., 2022; Ferri & Pesic, 2025; Gómez Fernández-Aguado et al., 2022; Gruenberg, 2024; Tümmler, 2022).
This study is structured as follows. Section 2 provides a review of the relevant literature on DIS and its impact on the banking stability in different economic conditions. Section 3 presents the research methodology, while Section 4 describes the data set used. Section 5 is dedicated to the analysis and interpretation of the empirical results, whereas the final section presents conclusions and discusses the theoretical and practical implications of the research.
2. Literature review
Scientific research consistently shows that the effects of DIS are not unambiguous, but depend on economic conditions. In stable periods, the system often encourages moral hazard and greater risk-taking (Angkinand & Wihlborg, 2010; Brandao-Marques et al., 2013; Chesini, 2014; Chiaramonte et al., 2020; Cull et al., 2005; Dewenter et al., 2018; González, 2005; Ioannidou & Penas, 2010; Kusairi et al., 2018; Lakštutienė et al., 2011; Laeven & Levine, 2009; Ngalawa et al., 2016; Ntwaepelo, 2023; Osborne & Lee, 2001; Prabha & Wihlborg, 2014; Yilmaz & Muslumov, 2008). Conversely, in crisis periods, the DIS helps stabilise the banking system by preventing depositor panic and large-scale withdrawals (Anginer et al., 2014; Engineer et al., 2013; Hasan et al., 2017; Prabha & Wihlborg, 2014; Weiß et al., 2014).
Although Anginer et al. (2014) also emphasise that the DIS weakens market discipline in calm periods and encourages banks to take on greater risk, they also note that, in crisis conditions, it contributes to preserving stability by reducing the risk of sudden deposit outflows. The authors further point out that the negative effects in stable conditions are often stronger than the positive effects in a crisis. Similarly, Prabha and Wihlborg (2014) argue that the DIS reduces overall risk for banks in crisis periods, but, in stable conditions, it encourages risky placements and thus can increase the vulnerability of the system in the long term.
Osborne and Lee (2001) warn that negative effects in stable periods can be a trigger for future crises. In their analysis, Demirgüç-Kunt and Detragiache (2002) show that systems with a higher coverage and lower level of supervision have higher probability of banking crises. Laeven (2002) adds that the DIS reduces depositor discipline, especially in banks with lower capital, which, in turn, increases business risk. Cull et al. (2005) and González (2005) emphasise that the problem of moral hazard intensifies if the system does not differentiate premiums according to the risk taken by banks, while Wagster (2007) emphasises the connection between extended deposit insurance and an increased propensity toward risky loans. Laeven and Levine (2009) empirically confirm that banks in countries with more developed and stricter DIS are less dependent on market discipline, which increases their vulnerability to future crises.
Furthermore, Yilmaz and Muslumov (2008) and Ioannidou and Penas (2010) find that the safety of depositors weakens their motivation to supervise banks, whereas banks use the existence of a DIS as a protection against the failure of risky investments. Angkinand and Wihlborg (2010) emphasise the importance of system design (e.g. risk-sensitive premiums) in mitigating negative effects. DeLong and Saunders (2011) indicate that the lack of transparency and information asymmetry between regulators, banks and depositors further aggravates the problem of moral hazard.
Lakštutienė et al. (2011) show that the negative effects are more pronounced in countries with weaker regulatory frameworks and institutional infrastructure. Brandao-Marques et al. (2013) and Chesini (2014) confirm that the DIS can increase systemic risk in the long term if it is not accompanied by strong supervisory mechanisms. Finally, Barth et al. (2006) and Ngalawa et al. (2016) emphasise that system design, regulatory framework and institutional context decisively determine the ultimate effects of deposit insurance.
Despite the extensive literature on deposit insurance and banking stability, existing studies predominantly focus on either cross-country comparisons or single-period analyses, without explicitly distinguishing between different phases of the economic cycle. Furthermore, limited attention has been given to comparative analyses between EU and SEE countries, particularly in the context of the Global Financial Crisis.
This study aims to address this gap by providing a dynamic panel analysis of the effects of deposit insurance across pre-crisis, crisis, and post-crisis periods, using both z-score and NPL as complementary measures of bank risk. In doing so, it contributes to the literature by offering a more nuanced understanding of the cyclical implications of deposit insurance systems in different institutional environments.
Based on the above research review and the identified research gap, the following research questions are formulated:
RQ1: How does the deposit insurance system affect the level of bank risk before, during, and after the GFC?
RQ2: To what extent do the effects of the deposit insurance system differ across the phases of the economic cycle?
In order to provide an empirical answer to these questions, the following hypotheses are proposed. To ensure consistency between the research design and the empirical framework, the general hypothesis is further decomposed into three sub-hypotheses reflecting the distinct phases of the economic cycle analysed in this study.
H1: The effect of the deposit insurance system on bank risk differs depending on the phase of the economic cycle.
H1a: In the pre-crisis period, the deposit insurance system is associated with higher incentives for bank risk-taking, reflecting moral hazard, adverse selection, and weaker market discipline.
This expectation is grounded in the literature which suggests that deposit insurance weakens market discipline and encourages banks to take on additional risk in stable periods (Anginer et al., 2014; Demirgüç-Kunt & Detragiache, 2002; Laeven, 2002)
H1b: During the crisis period, the deposit insurance system is associated with a stabilising effect on bank risk through increased depositor confidence and reduced likelihood of systemic instability.
Contrary to the previous hypothesis, during crisis periods, deposit insurance plays a key stabilising role by preventing depositor panic and large withdrawals (Anginer et al., 2014; Hasan et al., 2017).
H1c: In the post-crisis period, the deposit insurance system is associated with changes in bank risk, with effects that may differ depending on the risk measure used.
In the post-crisis period, the effects of deposit insurance remain ambiguous, as stabilisation mechanisms coexist with renewed incentives for risk-taking (Prabha & Wihlborg, 2014).
The above hypotheses are empirically tested using a dynamic panel data approach, which allows the assessment of the impact of deposit insurance system characteristics on bank risk across different phases of the economic cycle.
3. Methodology
This study uses a dynamic panel data approach to examine the effects of deposit insurance systems (DIS) on bank risk across different phases of the economic cycle. The empirical analysis employs the two-step Arellano-Bond difference Generalised Method of Moments (GMM) estimator developed by Arellano and Bond (1991).
A dynamic panel framework is appropriate because bank risk is persistent over time, with current risk levels influenced by previous conditions. Therefore, lagged values of the dependent variable are included among the regressors. In these circumstances, conventional panel estimators may yield biased and inconsistent results due to unobserved heterogeneity, autocorrelation, and potential endogeneity between bank risk and the explanatory variables.
The Arellano-Bond difference GMM estimator addresses these issues by transforming the model into first differences and using lagged levels of endogenous variables as internal instruments. This approach helps mitigate endogeneity arising from reverse causality, omitted variable bias, and the dynamic structure of the model. In line with the recommendations of Roodman (2009), the number of instruments is restricted to avoid instrument proliferation and preserve the validity of the Hansen test.
The two-step estimator is used because it provides more efficient and consistent estimates in the presence of heteroscedasticity. In the second step, residuals from the first-step estimation are used to construct a consistent variance-covariance matrix (Adeboye & Ogunnusi, 2026; Višić & Škrabić Perić, 2011). As standard two-step GMM standard errors may be downward biased in finite samples, robust standard errors corrected according to the methodology proposed by Windmeijer (2005) are applied.
Model validity and specification are evaluated using several diagnostic tests. First, the Arellano-Bond autocorrelation test is applied to examine serial correlation in the differenced residuals. Consistent with model assumptions, first-order autocorrelation is expected, while the absence of second-order autocorrelation indicates correct model specification (Arellano & Bond, 1991).
Second, the Hansen test of over-identifying restrictions is used to assess the validity of the instrumental variables. Compared with the traditional Sargan test, the Hansen test is more robust in the presence of heteroscedasticity and is therefore considered more appropriate for the estimated models. The null hypothesis of the Hansen test assumes that the instruments are valid and uncorrelated with the error term.
Finally, the Wald test is used to examine the joint statistical significance of the explanatory variables included in the models. The combination of these diagnostic tests supports the adequacy and reliability of the estimated specifications.
The empirical analysis is conducted using the STATA software package.
3.1. Sample and model
The sample covers 34 banking systems and 1,453 credit institutions over the period 2005–2014. The data are obtained from the Orbis Bank Focus database and include EU (1,330 credit institutions) and SEE countries (123 credit institutions). The sample structure reflects the focus of this study on comparing EU Member States with non-EU SEE countries, a group that has received relatively limited attention in the existing literature. By including both groups, this study aims to address this research gap and provide a more comprehensive analysis of deposit insurance systems across different institutional environments.
Given the research hypotheses and the absence of a universally accepted definition of bank stability, it is approximated in this research by a combination of bank risk indicators. It is operationalized through two dependent variables that are common in the relevant literature (Angkinand & Wihlborg, 2006, 2010; Anginer et al., 2014; Fang et al., 2014; Forssbaeck, 2011; Fonseca & González, 2010; González, 2005; Ioannidou & Penas, 2010; Lambert et al., 2017; Laeven & Levine, 2009; Lè, 2013; Yan et al., 2014).
1. Z-score, which measures the bank’s distance from insolvency (a higher score implies a lower risk appetite).
2. Share of non-performing loans (NPL) in total loans, which reflects the quality of the loan portfolio and exposure to credit risk.
In order to test the hypothesis, set out in this paper, according to which the effects of the deposit insurance system on banks’ risk-taking behaviour depend on the level of general economic activity, the following models were formulated:
z – scorei,t = a + yz – scorei,t–1 + β1DIS * BEFORECRISISi,t + + β2DIS * CRISISi,t + β3DIS * POSTCRISISi,t + β4lnFDPPCAPITAi,t + + β5SAVGDPi,t + β11EUSTATEi,t + εi,t; i = 1,…,1453, t = 1,…,10. (1)
NPLi,t = a + yNPLi,t–1+ β1DIS * BEFORECRISISi,t + β2DIS * CRISISi,t + + β3DIS * POSTCRISISi,t + β4lnFDPPCAPITAi,t + β5SAVGDPi,t+ + β11EUSTATEi,t + εi,t; i = 1,…,1453, t = 1,…,10. (2)
To examine the implications of the deposit insurance system on bank risk across different phases of the economic cycle, the key explanatory variables are specified as interaction terms between period-specific dummy variables and the deposit insurance variable (DIS). In the empirical model, DIS is operationalised as the natural logarithm of the deposit insurance coverage amount.
Specifically, three interaction variables are constructed: (DIS x PRECRISIS) for the pre-crisis period (2005–2007), (DIS x CRISIS) for the crisis period (2008–2012), and (DIS x POSTCRISIS) for the post-crisis period (2013–2014). These variables capture the period-specific effects of deposit insurance on bank risk, allowing the impact of deposit insurance to vary across different phases of the economic cycle. Definitions and descriptions of all variables included in the empirical models are reported in Table 4 in the Appendix.
In addition, dummy variables (PRECRISIS, CRISIS, POSTCRISIS) are included in alternative model specifications to account for the direct effects of macroeconomic conditions on bank risk. It is important to distinguish between these dummy variables, which reflect general economic conditions, and the interaction terms, which measure how the effect of deposit insurance changes across different periods. This specification follows the empirical approach of Anginer et al. (2014) and Prabha and Wihlborg (2014), enabling a more nuanced assessment of the cyclical implications of deposit insurance systems. However, unlike these studies, which define the crisis period as 2007–2009 due to their broader global sample, this study identifies the crisis period as 2008–2012 based on macroeconomic indicators such as GDP growth and GDP per capita within the observed countries.
In addition to the baseline specification, further model variations are estimated to address potential multicollinearity and to test the robustness of the results. These alternative specifications include additional characteristics of the deposit insurance system, such as the ratio of coverage amount to GDP per capita, the mode of system administration, and the type of premium, as well as macroeconomic control variables. Dummy variables are also included to indicate the pre-crisis, crisis, and post-crisis periods. The models are estimated using z-score and NPL as dependent variables.
z – scorei,t = α + γz – scorei,t–1+ β1PRECRISISi,t + β2limgdpi,t + + β3PREMIAi,t + β4ADMINISTRATIONi,t + β5lnGDPPCAPITAi,t + + β11SAVGDPi,t + β12EUSTATEi,t + εit; i = 1,...,1453, t = 1,…,10. (3)
z – scorei,t = α + γz – scorei,t–1+ β1CRISISi,t + β2limgdpi,t + β3PREMIAi,t + + β4ADMINISTRATIONi,t + β5lnGDPPCAPITAi,t + β11SAVGDPi,t + + β12EUSTATEi,t + εi,t; i = 1,...,1453, t = 1,…,10. (4)
z – scorei,t = α + γz – scorei,t–1+ β1POSTCRISISi,t + β2limgdpi,t + + β3PREMIAi,t + β4ADMINISTRATIONi,t + β5lnGDPPCAPITAi,t + + β11SAVGDPi,t + β12EUSTATEi,t + εi,t; i = 1,...,1453, t = 1,…,10. (5)
NPLi,t = α + γNPLi,t–1+ β1PRECRISISi,t + β2limgdpi,t + β3PREMIAi,t + + β4ADMINISTRATIONi,t + β5lnGDPPCAPITAi,t + β11SAVGDPi,t + + β12EUSTATEi,t + εi,t; i = 1,...,1453, t = 1,…,10. (6)
NPLi,t = α + γNPLi,t–1+ β1CRISISi,t + β2limgdpi,t + β3PREMIAi,t + + β4ADMINISTRATIONi,t + β5lnGDPPCAPITAi,t + β11SAVGDPi,t + + β12EUSTATEi,t + εi,t; i = 1,...,1453, t = 1,…,10. (7)
NPLi,t = α +γNPLi,t–1+ β1POSTCRISISi,t + β2limgdpi,t + β3PREMIAi,t + + β4ADMINISTRATIONi,t + β5lnGDPPCAPITAi,t + β11SAVGDPi,t + + β12EUSTATEi,t + εi,t; i = 1,...,1453, t = 1,…, 10. (8)
The period encompassing the GFC is particularly relevant for investigating the impact of DIS, as this crisis exposed the structural weaknesses of banking systems and regulatory frameworks at the time. In such conditions, the DIS serves as a key mechanism for limiting bank panics and preserving confidence. The crisis period therefore provides insight into the stabilising role of DIS during financial instability, while the pre- and post-crisis periods highlight potential moral hazard and risk-taking incentives associated with the financial safety net. This sample therefore enables a clearer comparison of the effects of DIS at different stages of the cycle and under varying financial conditions, which would not be possible in stable or single-phase periods.
In contrast, the same period also represents a limitation of this paper. Although the time frame 2005-2014 covers the pre-crisis, crisis, and post-crisis periods, it does not include subsequent regulatory changes, such as the full implementation of Basel III, the establishment of the SRM, and more recent reforms at EU level. Therefore, future research should focus on the period after 2014 to examine the effects of DIS in the new regulatory and institutional environment.
3.2. Descriptive statistics and correlation
Descriptive statistics show that the dummy variables DIS x PRECRISIS, DIS x CRISIS, and DIS x POSTCRISIS have lower standard deviations among the observed credit institutions. This is due to the varying number of institutions observed from countries included in the research sample and the differences in DIS characteristics across countries. Following the descriptive statistics, the results of the econometric model are preceded by an analysis of the correlation among the observed independent variables, as a high correlation between individual variables indicates multicollinearity, which distorts the significance of the estimated parameters. Pearson’s correlation coefficient ranges from -1 to 1 and indicates the direction and strength of the linear statistical correlation between variables. A value of 0 indicates no correlation between the observed variables, positive values indicate a consistent increase, and negative values indicate that as one variable increases, the other decreases. Complete correlation is observed at the extreme values of the coefficient. According to Field (2009) and Lovrić (2005), multicollinearity may be present when correlation coefficients between pairs of independent variables exceed 0.8 or are less than -0.8. The correlation matrix preceding the dynamic panel analysis is presented below.
Table 1a: Descriptive statistics for model variables with z-score and NPL as dependent variables
Source: Author’s calculations using STATA 13.0
Table 1b: Descriptive statistics for the deposit insurance system – related independent variables by general economic activity (pre-crisis, crisis, and post-crisis periods)
Source: Author’s calculations using STATA 13.0
The correlation matrix displays only correlation coefficients that are statistically significant at the 5% level (95% confidence level), with an asterisk indicating coefficients significant at the 1% level (99% confidence level). Based on the reported correlation coefficients, it is possible to identify pairs of independent variables that could potentially cause a multicollinearity problem. According to the adopted threshold, defined as a positive or negative correlation greater than 0.8, no variable exceeds the threshold of 0.8 or -0.8. The correlation between gross domestic product per capita (lnGDPPCAPITA) and EU membership (EUSTATE) is 0.7091, indicating a positive association between GDP per capita and EU membership; that is, higher GDP per capita is observed among EU members. The results for the variable representing non-EU membership (NOEUSTATE) are the same as those for the membership variable (EUSTATE), but with the opposite sign. Therefore, the correlation between NOEUSTATE and lnGDPPCAPITA is -0.7091. A positive correlation of 0.7159 was recorded between the GDS variable, which represents the interest rate on deposits adjusted for inflation, and the GDP per capita variable (lnGDPPCAPITA). Additionally, the correlation between lnGDPPCAPITA and SAVEGDP is 0.72, which implies that increases in GDP per capita are accompanied by increases in the savings rate. Among other pairs of independent variables, the correlation coefficients are not at levels that could lead to multicollinearity.
Table 2: Correlation matrix of dependent and independent model variables
Note: * indicates statistical significance at the level of 1%; bolded values suggest a correlation to 0.8.
** Empty cells in the correlation matrix do not indicate the absence of a relationship between variables, but rather correlations that are not statistically significant at the 5% level and are therefore not relevant for assessing multicollinearity in the model.
Source: Author’s calculations using STATA 13.0
The correlation matrix is followed by the estimation of the dynamic panel model using the two-step Arellano-Bond (AB) estimator.
4. Empirical data
This chapter presents the results of the dynamic panel analysis examining the implications of DISs for bank risk at different stages of the economic cycle. The empirical models are estimated using a two-step Arellano-Bond GMM estimator, which accounts for the dynamic nature of bank risk indicators and the potential endogeneity of the explanatory variables. Tables 3a to 3d present the estimated coefficients of the models, in which the dependent variables include the z-score and the share of non-performing loans in total loans (NPL). The models differ in how they incorporate the DIS characteristics: some specifications use interaction variables for the pre-crisis, crisis, and post-crisis periods, while alternative models use dummy variables indicating the phases of economic activity. The models also include variables for EU membership and non-membership to control for institutional differences across the observed countries.
Table 3a: Overview of the panel analysis results on the impact of deposit insurance systems on bank risk by general economic activity (including EU membership)
Note: *,**,*** – level of significance at 10%, 5% and 1%
Source: Author’s calculations using STATA 13.0
Table 3b: Overview of the panel analysis results on the impact of deposit insurance systems on bank risk by general economic activity (including EU membership)
Note: *,**,*** – level of significance at 10%, 5% and 1%
Source: Author’s calculations using STATA 13.0
Table 3c: Overview of the panel analysis results on the impact of deposit insurance systems on bank risk by general economic activity (including non-EU membership)
Note: *,**,*** – level of significance at 10%, 5% and 1%
Source: Author’s calculations using STATA 13.0
Table 3d: Overview of the panel analysis results on the impact of deposit insurance systems on bank risk by general economic activity (including non-EU membership)
Note: *,**,*** – level of significance at 10%, 5% and 1%
Source: Author’s calculations using STATA 13.0
The coefficients of the lagged dependent variable are positive and statistically significant in all model specifications, confirming inertia in the movement of bank risk indicators and supporting the use of the dynamic panel approach.
Variables describing the DIS characteristics are statistically significant in presented models, with the intensity and direction of the estimated effects varying according to the dependent variable and the economic period observed. The models also include macroeconomic control variables, such as gross domestic product per capita (lnGDPPCAPITA), the share of savings in GDP (SAVGDP), and the ratio of the amount of protection to GDP per capita (limGDP), all of which have statistically significant coefficients in several specifications. Institutional characteristics of DIS, including the administration of the system (ADMINISTRATION), the premium structure (PREMIA), and variables indicating membership or non-membership in the EU (EUSTATE and NOEUSTATE), are included to capture differences in the regulatory and institutional environment across the sample countries.
The Arellano–Bond test confirms the presence of first-order autocorrelation and the absence of second-order autocorrelation, supporting the correct specification of the model. The Wald test indicates the joint statistical significance of the explanatory variables.
A detailed interpretation of the results obtained and their connection with theoretical assumptions and relevant empirical literature is presented in the next section.
5. Results and discussion
The results of the panel analysis of the implications of DIS on bank risk in different economic periods show differentiated effects. It is important to note that the interpretation of coefficients differs depending on the risk measure used. An increase in the z-score indicates lower bank risk, while an increase in the share of non-performing loans (NPL) reflects higher credit risk. Therefore, the same coefficient sign may imply different risk implications across these two measures. The different behaviour of the z-score and NPL indicators may also reflect the fact that these variables capture different dimensions of bank risk. While the z-score is a broader forward-looking indicator of bank stability that incorporates profitability, leverage, and earnings volatility, NPL primarily reflects realised credit risk and tends to adjust more slowly to changes in macroeconomic conditions. Similar findings on the heterogeneous behaviour of financial risk indicators are reported by Svoboda et al. (2025), who show that different financial stability and risk assessment models may produce systematically different classifications depending on the risk dimension considered.
In the pre-crisis period (2005–2007), the DIS × PRECRISIS variable shows a positive and statistically significant relationship with the z-score, indicating a reduction in overall bank risk. In contrast, the relationship with NPL is negative but statistically insignificant, suggesting that credit risk dynamics were less sensitive to deposit insurance coverage during this period of economic expansion.
During the crisis period (2008–2012), the DIS × CRISIS variable exhibits a positive and significant relationship with the z-score, indicating a stabilising effect of deposit insurance on bank risk. However, the relationship with NPL is also positive and significant, reflecting an increase in credit risk. This finding suggests that deposit insurance primarily operates through the liability side of bank balance sheets by stabilising depositor behaviour and reducing liquidity pressures, with only a limited capacity to affect the quality of previously extended loans. These dynamics in bank stability are also illustrated in Graph 1 presented in the Appendix.
In the post-crisis period (2013–2014), the DIS × POSTCRISIS variable shows a positive and significant relationship with both the z-score and NPL. This suggests that while overall bank stability improves, as reflected in higher z-scores, credit risk remains elevated. These findings indicate that the stabilising effects of deposit insurance persist, but are accompanied by continued vulnerabilities in loan quality. The movement of NPL ratios across the observed period is presented in Graph 2 in the Appendix.
Overall, these results confirm that the effects of deposit insurance are not uniform across different dimensions of bank risk. While the system effectively reduces insolvency risk during periods of financial stress, it does not fully eliminate incentives for excessive risk-taking nor does it prevent the accumulation of credit risk, particularly after crisis periods. The findings therefore support the moral hazard hypothesis, which suggests that stronger deposit protection may weaken market discipline and encourage greater risk-taking during periods of economic stability.
Control variables such as Gross Domestic Product per capita (lnGDPPCAPITA) show a positive and significant relationship with the z-score and a negative and significant relationship with NPL, confirming that a higher level of economic development reduces bank risk. The savings ratio (SAVGDP) is positive and significant only with the z-score, meaning that an increase in savings reduces risk, primarily in periods of stability. EU membership (EUSTATE) is associated with higher risk, while non-membership (NOEUSTATE) shows the opposite effect, partly reflecting different experiences of banking crises within the EU. The ADMINISTRATION variable suggests that private DIS reduce risk, which is consistent with empirical research indicating that private models can have a disciplining effect ((Brandao-Marques et al., 2013; Calomiris, 1990; Demirgüç-Kunt & Detragiache, 2002; Demirgüç-Kunt et al., 2007; Garcia, 1996, 1999). The relationship between deposit insurance premiums (PREMIA) and risk measures is also positive and significant, confirming that more fairly structured premiums can reduce risk measured by the z-score, but at the same time increase exposure to credit risk (NPL).
The results indicate that the DIS contributed to reducing bank risk during the crisis period, as reflected in the higher z-score values, thereby supporting its stabilising function. However, the same system could not prevent the increase in NPLs, which are the result of the pre-crisis credit expansion and macroeconomic deterioration. This suggests that the stabilising effect of deposit insurance may be insufficient to fully offset the long-term consequences of excessive credit growth. From a regulatory perspective, these findings imply that deposit insurance alone cannot guarantee banking stability unless accompanied by effective supervision, prudent lending standards, and mechanisms limiting excessive risk-taking. Consistent with the research of Anginer et al. (2014), Anginer and Demirgüç-Kunt (2018), and Prabha and Wihlborg (2014), these findings support the hypothesis that the implications of deposit insurance schemes for bank risk vary across phases of the economic cycle. Overall, the findings provide empirical support for the hypothesis that the effects of deposit insurance are cyclical and depend on the broader macroeconomic environment, as well as on the specific dimension of bank risk being examined.
6. Conclusions
The aim of this paper was to examine the effects of DIS on bank risk across different phases of the economic cycle, with particular emphasis on the periods before, during, and after the GFC. Using a dynamic panel analysis on a sample of EU and SEE countries for the period 2005 to 2014, the relationship between the characteristics of DIS and two measures of bank risk, the z-score and the share of non-performing loans in total loans, was analysed. The results confirm that the effects of DIS on bank risk are not unambiguous, but largely depend on the phase of the economic cycle. During the financial crisis, the DIS played a pronounced stabilising role, reflected in reduced risk as measured by the z-score. This finding highlights the importance of DIS in preventing depositor panic and maintaining confidence in the banking system during periods of increased financial uncertainty.
In contrast, the results also indicate the limited capacity of the DIS to influence banks’ credit risk. During the crisis and in the post-crisis period, the increase in the level of deposit protection did not prevent the increase in non-performing loans, confirming that credit risk is largely a consequence of pre-crisis credit expansion and deteriorating macroeconomic conditions. This demonstrates that, although a DIS is crucial for maintaining the stability of the deposit base, it cannot independently address the structural weaknesses of banks’ credit portfolios.
During the pre-crisis and post-crisis periods, increased risk appetite was observed, indicating the presence of moral hazard and weaker market discipline in a stable economic environment. Furthermore, the results highlight the importance of the design of DIS, with features such as the level of protection relative to GDP, the premium structure, and system administration exerting a significant influence on bank behaviour and the degree of risk assumed.
The findings of this study add to the existing literature by providing empirical evidence on the cyclical effects of DIS on bank risk in Europe. The main contribution of the paper is its comparative analysis of EU and SEE countries, highlighting differences in institutional and regulatory environments that influence the effectiveness of DIS.
The results have important implications for ongoing debates on further development of the Banking Union and the creation of a common EDIS. The findings indicate that an increase in the level of deposit protection can have positive effects during crises, while the long-term stability of the banking system requires a carefully designed scheme that protects depositors yet limits incentives for excessive risk-taking.
The limitations of this paper relate mainly to the time frame of the analysis, which ends in 2014 and does not cover subsequent regulatory changes, such as the full implementation of Basel III, the establishment of the Single Resolution Mechanism (SRM), and recent reforms at EU level. Future research should examine more recent periods to assess whether changes in the regulatory framework have altered the effects of DIS on bank risk, and provide a deeper analysis of the interaction between DIS, supervision, and other elements of the financial safety net. These findings support hypotheses H1a–H1c, confirming the cyclical nature of deposit insurance effects.
Note: This manuscript is based on the author’s doctoral thesis: “Deposit insurance system in the function of banking stability”.
Acknowledgement: This paper has been written within the scope of the project Regulation and Sustainability in Finance: Challenges and Opportunities for the Stability of the Financial System (uniri-iz-25-33), funded by European Union – NextGenerationEU via the Croatian National Recovery and Resilience Plan 2021-2026, in conjunction with the University of Rijeka, Faculty of Economics and Business Programme Financing. The views and opinions expressed are solely those of the authors and do not necessarily reflect the official position of the European Union or the European Commission. Neither the European Union nor the European Commission can be held responsible for them.
References
Adeboye, N. O., & Ogunnusi, N. O. (2026). Dynamic Panel Model Estimation in the Presence of Heteroscedasticity [Preprint]. SSRN. http://dx.doi.org/10.2139/ssrn.6096835
Alam, N., Sivarajah, G., & Bhatti, M. I. (2021). Do Deposit Insurance Systems Promote Banking Stability? International Journal of Financial Studies, 9(3), Article 52. https://doi.org/10.3390/ijfs9030052
Allen, F., Carletti, E., & Leonello, A. (2011). Deposit insurance and risk taking. Oxford Review of Economic Policy, 27(3), 464–478. https://doi.org/10.1093/oxrep/grr022
Anginer, D., Demirgüç-Kunt, A., & Zhu, M. (2014). How does deposit insurance affect bank risk? Evidence from the recent crisis. Journalof Banking & Finance, 48(C), 312 – 321.
Anginer, D., & Demirgüç-Kunt, A. (2018). Bank runs and moral hazard: A review of deposit insurance (Working Paper No. 5859). World Bank Group. https://doi.org/10.1596/1813-9450-8589
Angkinand, A. P., & Wihlborg, C. (2006). Deposit insurance coverage, ownership, and banks’ risk-taking in Europe (LEFIC Working Paper). Leuven Centre for Economics and Finance.
Angkinand, A., & Wihlborg, C. (2010). Deposit insurance coverage, ownership and banks’ risk-taking in emerging markets. Journal of International Money and Finance, 29(2), 252–274. https://doi.org/10.1016/j.jimonfin.2009.08.001
Arda, A., & Dobler, M. (2022). The Role for Deposit Insurance Funds in Dealing with Failing Banks in the European Union (Working Paper No. 22/2). International Monetary Fund. https://doi.org/10.5089/9781616358075.001
Arellano, M., & Bond, S. (1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment. The Review of Economic Studies, 58(2), 277–297. https://doi.org/10.2307/2297968
Ayadi, R., & Lastra, R. M. (2010). Proposals for reforming deposit guarantee schemes in Europe. Journal of Banking Regulation, 11(3), 210–222. https://doi.org/10.1057/jbr.2010.9
Barth, J. R., Caprio G., & Levine, R. (2006). Rethinking bank regulation: Till Angels govern. Cambridge University Press https://doi.org/10.1017/cbo9780511753817
Beck, T., Krahnen, J.-P., Martin, P., Mayer, F., Pisani-Ferry, J., Tröger, T., Weder di Mauro, B., Véron, N., & Zettelmeyer, J. (2022). Completing Europe’s banking union: Economic requirements and legal conditions (Policy Contribution No. 20/22). Brugel. https://www.bruegel.org/system/files/2022-11/PC%2020%202022_5.pdf
Brandao-Marques, L., Correa, R., & Sapriza, H. (2013). International evidence on government support and risk taking in the banking sector (International Finance Discussion Paper). Board of Governors of the Federal Reserve System. https://doi.org/10.17016/IFDP.2013.1086
Calomiris, C. W. (1990). Is Deposit Insurance Necessary? A Historical Perspective. The Journal of Economic History, 50(2), 283–295. https://doi.org/10.1017/S0022050700036433
Calomiris, C. W., & Jaremski, M. (2016). Deposit Insurance: Theories and Facts. Annual Review of Financial Economics, 8, 97–120. https://doi.org/10.1146/annurev-financial-111914-041923
Carisano, R. (1992). Deposit insurance: Theory, Policy and Evidence. Dartmouth Publishing Company Limited.
CESifo DICE (2008). Flexicurity (Report No. 4/2008). https://www.ifo.de/DocDL/CESifoDICEreport408.pdf
CESifo DICE (2011). Migration–challenges and policies (Report No. 4/2011). https://www.ifo.de/DocDL/CESifoDICEreport411.pdf
Chesini, G. (2014). Does deposit insurance increase moral hazard in banks? The case of the Australian Financial Claims Scheme. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.2485091
Chiaramonte, L., Girardone, C., Migliavacca, M., & Poli, F. (2020). Deposit insurance schemes and bank stability in Europe: how much does design matter? The European Journal of Finance, 26(7–8), 589–615. https://doi.org/10.1080/1351847X.2019.1607763
Cull, R., Senbet, L. W., & Sorge, M. (2005). Deposit Insurance and Financial Development. Journal of Money, Credit and Banking, 37(1), 43–82. http://www.jstor.org/stable/3838937
Cullen,J. (2023). Moral Hazard and the deposit insurance cap: a case study of Northern Rock and Silicon Valley Bank. Law and Financial Markets Review, 17(4), 272–297. https://doi.org/10.1080/17521440.2024.2403395
Dávila, E., & Goldstein, I. (2023). Optimal Deposit Insurance. Journal of Political Economy, 131(7), 1676–1730. https://doi.org/10.1086/723633
DeLong, G., & Saunders, A. (2011). Did the introduction of fixed-rate federal deposit insurance increase long-term bank risk-taking? Journal of Financial Stability, 7(1), 19–25. https://doi.org/10.1016/j.jfs.2008.09.013
Demirgüç-Kunt, A., & Detragiache, E. (2002). Does deposit insurance increase banking system stability? An empirical investigation. Journal of Monetary Economics, 49(7), 1373–1406. https://doi.org/10.1016/s0304-3932(02)00171-x
Demirgüç-Kunt, A., Kane, E. J., & Laeven, L. (2007). Determinants of Deposit-Insurance Adoption and Design (Working paper No. 12862). National Bureau of Economic Research. https://www.nber.org/system/files/working_papers/w12862/w12862.pdf
Demirgüç-Kunt, A., Kane, E. J., & Laeven, L. (2014). Deposit insurance database (IMF Working Paper No. 14/118). International Monetary Fund.
Dewenter, K. L., Hess, A. C., & Brogaard, J. (2018). Institutions and Deposit Insurance: Empirical Evidence. Journal of Financial Services Research, 54(3), 269–292. https://doi.org/10.1007/s10693-017-0271-8
Duttagupta, R., & Cashin, P. A. (2008). The Anatomy of Banking Crises (Working Paper No. 08/93). International Monetary Fund.https://doi.org/10.5089/9781451869545.001
Engineer, M. H., Schure, P., Gillis, M. (2013). A positive analysis of deposit insurance provision: Regulatory competition among European Union countries. Journalof Financial Stability, 9(4), 530–544. https://doi.org/10.1016/j.jfs.2013.10.001
European Council - Council of European Union (2024, June 19). Banking union. https://www.consilium.europa.eu/en/policies/banking-union/?utm
Fang, Y., Hasan, I., & Marton, K. (2014). Institutional development and bank stability: Evidence from transition countries, Journal of Banking & Finance, 39, 160–176.https://doi.org/10.1016/j.jbankfin.2013.11.003
Faulend, M. & Kraft, E. (2004). Hrvatski sustav osiguranja štednih uloga: treba li ga mijenjati, i ako treba, što promijeniti i kada? [Croatia’s deposit insurance system: should it be reformed, and if so, what should be changed and when?]. Financijska teorija i praksa, 28(1), 121–139.https://hrcak.srce.hr/5754
Faulend, M. (2002). Osiguranje štednih uloga [Deposit Insurance of Savings Deposits] [Master’s thesis, University of Zagreb]. Croris.https://www.croris.hr/crosbi/publikacija/ocjenski-rad/380839
Ferri, G., & Pesic, V. (2025). Can the banking union work without common deposit Guarantees? The initial market assessment and policy implications. International Review of Economics & Finance, 101, Article 104159, https://doi.org/10.1016/j.iref.2025.104159
Field, A. (2009). Discovering statistics using SPSS (3rd ed.). SAGE publications.
Fonseca, A. R., & González, F. (2010). How bank capital buffers vary across countries: The influence of cost of deposits, market power and bank regulation. Journal of Banking & Finance, 34(4), 892–902. https://doi.org/10.1016/j.jbankfin.2009.09.020
Forssbaeck, J. (2011). Ownership structure, market discipline and banks’ risk-taking incentives under deposit insurance. Journal of Banking & Finance, 35(10), 2666–2678.https://doi.org/10.1016/j.jbankfin.2011.02.024
Garcia, G. (1996). Deposit Insurance: Obtaining the Benefits and Avoiding the Pitfalls (Working Paper No. 1996/083). International Monetary Fund.https://doi.org/10.5089/9781451850499.001
Garcia, G. (1999). Deposit insurance: A Survey of Actual and Best Practices (Working Paper No. 1999/054). International Monetary Fund.https://doi.org/ 10.5089/9781451847499.001
Gómez Fernández-Aguado, P., Martínez, E. T., Ruíz, R. M., & Partal Ureña, A. (2022). Evaluation of European Deposit Insurance Scheme funding based on risk analysis. International Review of Economics & Finance, 78, 234–247. https://doi.org/10.1016/j.iref.2021.11.013
González, F. (2005). Bank regulation and risk-taking incentives: An international comparison of banks risk. Journal of Banking & Finance, 29(5), 1153–1184, https://doi.org/10.1016/j.jbankfin.2004.05.029
Gropp, R., & Vesala, J. (2004). Deposit insurance, moral hazard and market monitoring. Review of Finance, 8(4), 571–602. https://doi.org/10.1093/rof/8.4.571
Gruenberg, M. J. (2024, October 29). A tale of two unions: Deposit insurance in the United States and Europe (Speech). Federal Deposit Insurance Corporation. https://www.fdic.gov/news/speeches/2024/tale-two-unions-deposit-insurance-united-states-and-europe?source=govdelivery&utm_medium=email&utm_source=govdelivery
Hasan, I., Fang, Y., Liu, L., & Zhang, G. (2017). Deposit Insurance and the 2008-2009 Global Financial Crisis. Financial Stability Studies, 18(1), 1–19. https://doi.org/10.26588/kdic.2017.18.1.001
Huberdeau-Reid, D. A., & Pennacchi, G. G. (2025). Adverse selection in deposit insurance and government funding following the 2023 banking crisis. Journal of Financial Intermediation, 63, Article 101165. https://doi.org/10.1016/ j.jfi.2025.101165
Ioannidou, V. P., & Penas, M. F. (2010). Deposit insurance and banking risk-taking: evidence from internal loan rating. Journal of Financial Intermediation, 19(1), 95–115. https://doi.org/10.1016/j.jfi.2009.01.002
Kahn, C. M., & Santos, J. A. (2005). Allocating bank regulatory powers: Lender of last resort, deposit insurance and supervision. European Economic Review, 49(8), 2107–2136. https://doi.org/10.1016/j.euroecorev.2004.10.004
Kundid Novokmet, A. (2015). Kontroverze regulacije banaka kroz kapitalne zahtjeve [Controversies of bank capital requirements regulation]. Ekonomski pregled, 66(2), 156 – 176.
Kusairi, S., Sanusi, N. A., & Ismail, A. G. (2018). Dilemma of deposit insurance policy in ASEAN countries: Does it promote banking industry stability or moral hazard? Borsa Istanbul Review, 18(1), 33–40. https://doi.org/10.1016/j.bir.2017.08.006
Kuznichenko, P., Frolov, S., Orlov, V., & Boiko, O. (2021). European Deposit Insurance Scheme implementation: pros and cons. Banks and Bank Systems, 16(1), 116–126. https://doi.org/10.21511/bbs.16(1).2021.11
Laeven, L. (2002). Bank risk and deposit insurance. The World Bank Economic Review, 16(1), 109–137. https://doi.org/10.1093/wber/16.1.109
Laeven, L., & Levine, R. (2009). Bank governance, regulation and risk taking. Journal of Financial Economics, 93(2), 259–275, https://doi.org/10.1016/j.jfineco.2008.09.003
Laeven, L., & Valencia, F. (2013). Systemic Banking Crises Database. IMF Economic Review, 61(2), 225–270. https://doi.org/10.1057/imfer.2013.12
Lakštutienė, A., Krušinskas, R., & Rumšaitė, D. (2011). The influence of deposit insurance on the stability of the Baltic States banking system. Journal of Business Economics and Management, 12(3), 482–502. https://doi.org/10.3846/16111699.2011.599413
Lé, M. (2013). Deposit insurance adoption and bank risk-taking: The role of leverage (PSE Working Paper No. 2013–41). Paris School of Economics. https://doi.org/10.2139/ssrn.2363376
Lambert, C., Noth., F., & Schüwer, U. (2017). How do insured deposits affect bank risk? Evidence from the 2008 Emergency Economic Stabilization Act. Journal of Financial Intermediation, 29, 81–102. https://doi.org/10.1016/j.jfi.2015.07.001
Lovrić, Lj. (2005). Uvod u ekonometriju. Sveučilište u Rijeci, Ekonomski fakultet.
Martin, C., Puri, M., & Ufier, A. (2016). On deposit stability in failing banks [Preliminary draft]. SSRN. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2898425
Matutes, C., & Vives, X. (1996). Competition for Deposits, Fragility, and Insurance. Journal of Financial Intermediation, 5(2), 184–216. https://doi.org/10.1006/jfin.1996.0010
Merton, R. C. (1977). An analytic derivation of the cost of deposit insurance and loan guarantees An application of modern option pricing theory. Journal of Banking & Finance, 1(1), 3–11. https://doi.org/10.1016/0378-4266(77)90015-2
Ngalawa, H., Tchana, F. T., & Viegi, N. (2016). Banking Instability and Deposit Insurance: The Role of Moral Hazard. Journal of Applied Economics, 19(2), 323–350. https://doi.org/10.1016/s1514-0326(16)30013-7
Ntwaepelo, K. M. (2023). Bank Stability versus Financial Development: A Generous Deposit Insurer’s Dilemma (Discussion Paper No. 2023-09). Department of Economics, University of Reading. https://research.reading.ac.uk/economics/wp-content/uploads/sites/87/2023/07/emdp202309.pdf
Orbis Bank Focus Database. (2017, January 20). https://www.moodys.com/web/en/us/capabilities/company-reference-data/orbis.html
Osborne, D. K., & Lee, S. (2001). Effects of Deposit Insurance Reform on Moral Hazard in US Banking. Journal of Business Finance & Accounting, 28(7-8), 979–992. https://doi.org/10.1111/1468-5957.00401
Pierret, L., & Howarth, D. (2023). Moral Hazard, central bankers, and Banking Union: professional dissensus and the politics of European financial system stability. Journal of European Integration, 45(1), 15–41. https://doi.org/10.1080/07036337.2022.2156501
Prabha, A. P., Wihlborg, C. (2014). Implicit Guarantees, business models and banks’ risk-taking through the crisis: Global and European perspectives. Journal ofEconomics and Business, 76, 10–38. https://doi.org/10.1016/j.jeconbus.2014.01.001
Protić, M. (2002). Osiguranje depozita i problem moralnog hazarda (primer nemačke šeme privatnog osiguranja depozita) [Deposit Insurance and the Problem of Moral Hazard (The Case of the German Private Deposit Insurance Scheme)]. Industrija, (1-4), 23–32. https://scindeks-clanci.ceon.rs/data/pdf/0350-0373/2002/0350-03730201023P.pdf
Roodman, D. (2009). How to Do Xtabond2: An Introduction to Difference and System GMM in Stata. The Stata Journal, 9(1), 86–136, https://doi.org/10.1177/1536867X0900900106
Schich, S. (2008). Financial crisis: deposit insurance and related financial safety net aspects. OECD Journal Financial Market Trends, 2008/2(95), 73–111. https://doi.org/10.1787/fmt-v2008-art12-en
Single Resolution Board (2025, September 18). Towards a simpler, stronger Banking Union. The Eurofi Views Magazine. https://www.srb.europa.eu/en/content/towards-simpler-stronger-banking-union-eurofi-article-srb-chair-dominique-laboureix?utm
Suljić Nikolaj, S., Olgić Draženović, B., & Buterin, V. (2022). Deposit insurance, banking stability and banking indicators. Economic research-Ekonomska istraživanja, 35(1), 5632–5649. https://doi.org/10.1080/1331677X.2022.2033130
Svoboda, J., Berková, I., Pražáková, J., & Vejsadová Dryjová, M. (2025). Relationships Between the Beneish M-Score and Bankruptcy Models: Insights from Financial Models in Czech Manufacturing Firms. Zbornik radova Ekonomskog fakulteta u Rijeci [Proceedings of Rijeka Faculty of Economics: Journal of Economics and Business], 43(2), 465–484. https://doi.org/10.18045/zbefri.2025.2.7
Tümmler, M. (2022). Completing Banking Union? The Role of National Deposit Guarantee Schemes in Shifting Member States’ Preferences on the European Deposit Insurance Scheme. Journal of Common Market Studies, 60(6), 1556–1572. https://doi.org/10.1111/jcms.13318
Učkar, D., & Petrović, D. (2021). Efficiency of banks in Croatia. Zbornik Radova Ekonomskog Fakulteta U Rijeci, 39(2), 349–379.https://doi.org/10.18045/zbefri.2021.2.349
Višić, J., & Škrabić Perić, B. (2011). The determinants of value of incoming cross-border mergers & acquisitions in European transition countries. Communist and Post-Communist Studies, 44(3), 173–182. https://doi.org/10.1016/j.postcomstud.2011.07.004
Wagster, J. D. (2007). Wealth and Risk Effects of Adopting Deposit Insurance in Canada: Evidence of Risk Shifting by Banks and Trust Companies. Journal of Money, Credit and Banking, 39(7), 1651–1681. https://doi.org/10.1111/ j.1538-4616.2007.00082.x
Weiß, G. N. F., Bostandzic, D., & Neumann, S. (2014). What factors drive systemic risk during international financial crises? Journal of Banking & Finance, 41, 78–96. https://doi.org/10.1016/j.jbankfin.2014.01.001
Wheelock, D. C., & Kumbhakar, S. C. (1995). Which Banks Choose Deposit Insurance? Evidence of Adverse Selection and Moral Hazard in a Voluntary Insurance System. Journal of Money, Credit and Banking, 27(1), 186–201. https://ideas.repec.org/a/mcb/jmoncb/v27y1995i1p186-201.html
Windmeijer, F. (2005). A finite sample correction for the variance of linear efficient two-step GMM estimators. Journal of Econometrics, 126(1), 25–51. https://doi.org/10.1016/j.jeconom.2004.02.005
World Bank. (2026a). World Development Indicators: GDP per capita (current US$) (NY.GDP.PCAP.CD) [Data set]. World Bank Group. https://data.worldbank.org/indicator/NY.GDP.PCAP.CD
World Bank. (2026b). World Development Indicators: Gross savings (% of GDP) (NY.GDS.TOTL.CD) [Data set]. World Bank Group. https://data.worldbank.org/indicator/NY.GDS.TOTL.CD
Yan, X., Skully, M., Avram, K., & Vu, T. (2014). Market Discipline and Deposit Guarantee: Evidence from Australian Banks. International Review of Finance, 14(3), 431–457. https://doi.org/10.1111/irfi.12028
Yilmaz, E., & Muslumov, A. (2008). Deposit insurance and moral hazard problem: the case of Turkish banking system. Applied Economics, 40(16), 2147–2163. https://doi.org/10.1080/00036840600949306
Zolea, R. (2023). A Functional Analysis of the Role of Deposits in the Traditional Banking Industry. Review of Political Economy, 35(4), 933–952.https://doi.org/10.1080/09538259.2023.2233870
Sustav osiguranja depozita u različitim financijskim uvjetima
Stella Suljić Nikolaj2
