Tóm tắt Luận án Impact of monetary policy and macroprudential policy on bank stability in Vietnam

l stability and smoother economic and financial cycles, price stability as well as specific industrial policies. With to European Central Bank (ECB), the goals of macroprudential policy including (i) prevent the inordinate building-up of risk, resulting from out extraneous factor and market breakdown, to smoothen the financial cycle; (ii) make the financial sector more resilient and limit contagion effects; (iii) encourage 7 a system-wide perspective in financial regulation to create the right set of incentives for market participants. 2.3. BANK STABILITY 2.3.1. Bank stability concept Heretofore, several studies mentioned bank stability, but its definition remains debatable. Analyzing bank stability, the studies were researched in two directions, the first is financial stability researches related banking sector and the second is bank instability as an indirect approach to assessing bank stability, bank instability is the reverse state of bank stability. Based on previous studies, bank stability is a state in which not only banks can operate and perform their functions smoothly but also withstanding shocks from the outside environment. The banks themselves do not cause negative shocks to effect on economy so that they could contribute to developing financial system in particular and economy in general. 2.3.2. The role of bank stability Not only the operation of banks but also the operation of non-banks institutions can be smooth and effective when banks develop stably and sustainably. This is an important factor for government to easily manage macroeconomy, inflation, increasing people’s income. As a result, it enhances the competitiveness of the banking sector in particular and the nation in general on the international market. 2.4. THE IMPACT OF MONETARY AND MACROPRUDENTIAL POLICY ON BANK STABILITY 2.4.1. The impact of monetary policy on bank stability According to Madura (2014), monetary policy has a strong impact on interest rates, economic growth, so it affects the valuation of most assets in the financial market. Especially, 8 monetary policy affects commercial banks' operation through three main markets: money market, bond market, and mortgage market (Bernanke & Blinder, 1992). It impacts on interest rates or bond’s issuance and trading in the money and bond market. In the mortgage market, monetary policy effects on housing demand, debt financing market, interest rates on new loans, and risk premium for a mortgage (Madura, 2014). 2.4.2. Impact of macroprudential policy on bank stability According to IMF (2013), the macroprudential policy as a policy that uses prudential tools to limit financial system risks. In fact, macroprudential policy has been used successfully in some emerging economies before the global financial crisis and previous crisis periods. Otherwise, the financial system consists of 4 components: financial markets, financial institutions, financial instruments, and financial infrastructure. Commercial banks are one of the financial institutions along with the securities commission, insurance companies, financial leasing companies, and other types of banks. For these reasons, the macroprudential policy as a policy that uses of prudential tools to curb system risks to stabilize financial stability and bank stability. 2.4.3. Impact of monetary and macroprudential policy on bank stability According to Mishkin (2012), the credit channel is one of the transmission channels of monetary policy, so the banking system is an important subject in its transmission. State and effective banking system is the key to improving the effectiveness of monetary policy and reducing unexpected risks due to the process of regulating the money supply. At that time, macroprudential policy has the role of identifying potential risks, issuing warnings, implementing measures to ensure safety, preventing excessive risk-taking behavior, and coordinating with monetary policy to propose appropriate measures. The tools of macroprudential policy with the goal of financial stability prevent “distortion” and unexpected effects of monetary policy. For example, the impact of tightening monetary policy can be used by DTI ratio (IMF, 2013). Conversely, when monetary is relaxed to push assets price, LTV ratio can be used to reduce economic vulnerabilities. The tightening 9 of leverage ratios or liquidity ratio can reduce banks' risks in the case of higher reserve requirements. (IMF, 2013). 2.5. RESEARCH HYPOTHESIS H1: Bank stability has a positive relationship with bank stability year ago. H2: Money supply M2 has a negative impact on bank stability. H3: Rediscount interest rate has a positive impact on bank stability. H4: The capital adequacy ratio correlates with bank stability positively. H5: The liquidity ratio has a positive impact on bank stability. H6: The loan deposit ratio correlates with bank stability negatively. H7: In the case, the SBV increases money supply M2 into the economy and simultaneously allows commercial banks to increase loan deposit ratio, the stability of commercial banks will increase. H8: Bank size correlates with bank stability positively. H9: The cost operation to income operation ratio has a negative relationship with bank stability. H10: The loan to total assets ratio correlates with bank stability negatively. H11: The GDP has a positive impact on bank stability H12: Inflation has a negative impact on bank stability. 2.6. OVERVIEW OF STUDY 2.6.1. Overview of study on bank stability The studies about bank stability (instability) or bank risks were focused on the following contents: Firstly, the studies involved factors effect on bank stability, bank instability, bank risks (see Dwumfour (2017), Čihák and Hesse (2010), Köhler (2015)). In Vietnam, there was Hà and Hướng (2016)’s research on bank bankruptcy risk. 10 Secondly, besides the studies determined factors affecting bank stability (bank instability, bank risks), the studies analyzed the relationship between competition-stability (for example Goetz (2018), Jayakumar, Pradhan, Dash, Maradana, and Gaurav (2018), Fernández, González, and Suárez (2016), Beck, De Jonghe, and Schepens (2013). Phan, Anwar, Alexander, and Phan (2019), Tuyền, Đạo, and Anh (2017)) or liquidity-stability (see Khan, Scheule, and Wu (2017), Wagner (2007)) or diversification – stability (Abuzayed, Al- Fayoumi, and Molyneux (2018)) 2.6.2. Overview of the study on the impact of monetary policy on bank stability The studies about the impact of monetary policy on the real economy through commercial banks have been concerned by many economists and policymakers, especially in regarding bank risks and bank stability. These studies focused on: Firstly, the studies researched the transmission mechanism of monetary policy such as Mishkin (1996), Anil K. Kashyap and Stein (1995); (Anil K Kashyap & Stein, 2000), Kishan and Opiela (2000). Secondly, the studies related to the impact of monetary policy on the banking system such as Berkelmans, Kelly, and Sadeghian (2016), Borio, Gambacorta, and Hofmann (2017). Nguyen Thanh, Huong Vu, and Thu Le (2017) researched the relationship between monetary policy and bank profit in Vietnam. Thirdly, the studies involved the impact of monetary policy on bank stability (bank instability). Most of these studies showed that monetary policy affects bank stability through three channels including interest rate channel, credit channel, and risk-taking channel. Conclusion, the previous studies such as De Nicolò, Dell'Ariccia, Laeven, and Valencia (2010), Dell'Ariccia, Marquez, and Laeven (2010), Angeloni, Faia, and Lo Duca (2015), de Moraes and de Mendonça (2019), Ha and Quyen (2018) showed that, when central banks implement expansionary monetary policy, increasing money supply into the economy which affects reducing interest rates and increasing financial leverage of the bank, thereby lead to bank instability. Contrariwise, expansionary monetary policy also reduces the adverse selection, so that bank stability increase. A question is, in Vietnam, how does monetary policy 11 effect on bank stability? Because most of the studies conducted in emerging and advanced economies. In Vietnam, most of the studies related to the impact of monetary policy on bank profit or bank performance. In the case of the studies related to bank risks, it was the relationship between monetary policy and bank competition. So that, it is necessary to study impact of monetary policy on bank stability in Vietnam to addition empirical evidence for policymaker to have appropriate policy implications. 2.6.3. Overview of the study on the impact of macroprudential policy on bank stability One of the first studies on the effectiveness of macroprudential tools was Lim et al. (2011). Dell’Ariccia et al. (2012) analyzed the relationship between macroprudential policy and financial crisis riks, Claessens, Ghosh, and Mihet (2013) studied macroprudential policy to mitigate financial system vulnerabilities. Kuttner and Shim (2016) studies the effectiveness of nine non-interest rate policies on house prices and housing credit. Zhang and Zoli (2016) studied assessing effectiveness of macroprudential tools on credit and asset price cycles. Lee, Asuncion, and Kim (2016) analyzed how effective macroprudential policies control credit growth, leverage growth, and housing price appreciation. Aiyar, Calomiris, and Wieladek (2016) studied the interaction of monetary policy and capital requirement regulation of UK banks’. Fendoğlu (2017) assessed the effectiveness of macroprudential policy tools in containing credit cycles in major emerging market economies. Olszak, Roszkowska, and Kowalska (2018) analyze the effectiveness of various macroprudential policy instruments in reducing the procyclicality of loan-loss provision (LLPs). Several studies analyzed the effectiveness of macroprudential policy such as Akinci and Olmstead-Rumsey (2018), Cerutti, Claessens, and Laeven (2017),. Altunbas et al. (2018) investigated the effects of macroprudential policies on bank risk in 61 advanced and emerging market economies. In Vietnam, there were a few of studies on impacts of macroprudential policy on financial stability such as Trần Thị Kim Oanh et al. (2017), Vũ Hải Yến và Trần Thanh Ngân (2016). In conclusion, the studies on macroprudential policy focused on a relationship with financial stability, there were a few studies impact of macroprudential policy on bank risks, 12 but these studies researched in countries outside Vietnam. So the thesis impact of macroprudential policy on bank stability is necessary to add empirical evidence in Vietnam. 2.6.4. Overview of the study on the impact of monetary and macroprudential policies on bank stability So far, there have been several studies on the interaction between monetary and macroprudential policies on maintaining financial stability in general and bank stability in particular. Malovaná and Frait (2017) analyzed the relationship between monetary and macroprudential policy (interaction or conflict). Bruno, Shim, and Shin (2017) provided a comparative assessment of the effectiveness of macroprudential policy in 12 Asia-Pacific economies during the 2004-2013 period. Maddaloni and Peydró (2013) studied the impact of monetary and macroprudential policies on bank stability through the credit condition of 17 European banks in the period from Q4/2002 to Q4/2010. In Vietnam, Nguyễn Đức Trung và Nguyễn Hoàng Chung (2018) analyzed the impact of monetary and macroprudential policy on financial stability, Nguyễn Phi Lân et al. (2017) studies coordination between monetary policy and prudential policy for bank activities in Vietnam. 2.6.5. Research gaps Firstly, there has been a number of studies on the impact of monetary policy on bank stability, but these results were arguable whether the expansion or tightening monetary policy would maintain bank stability. Simultaneously, these studies have been conducted mainly in emerging or advanced economies. Secondly, the studies on macroprudential policy focused on (i) the impact of this policy in limiting systemic risks to reduce costs for the nation’s financial systems. There have even been researches on the effect of macroprudential policy on bank risks in the world’s nations (not including Vietnam). In Vietnam, most of researches analyzed the effectiveness of macroprudential policy and the impact of this policy on financial stability without emphasizing bank stability. 13 Thirdly, there have been researches on impact of monetary and macroprudential policies on bank stability, but this research studied for European countries and the authors used a credit channel to represent bank stability. From these above contents, the research gaps of the thesis are: Firstly, the thesis provides empirical evidence of the impact of monetary and macroprudential policy on stability of Vietnamese commercial banks to contribute an overview of the determinants of bank stability, in which, the variables of each policy show a negative and positive related to bank stability. Secondly, the thesis analyzed the impact of monetary policy and macroprudential policy on stability of 22 Vietnamese commercial banks in the period 2008-2018. In fact, how to operate these policies to both achieve the objectives and maintain bank stability is quite difficult. The research focuses on the impact of each policy on bank stability to suggest policy implications for the SBV. Thirdly, besides considering the individual effects, the thesis also analyzes the interaction between monetary and macroprudential policies on stability of commercial banks. The thesis focuses on these above contents to provide important evidence for bank management and policymakers. On these bases, the thesis proposes solutions and policy implications to maintain bank stability and improve the effectiveness of these policies. CHAPTER 3 METHODOLOGY 3.1. RESEARCH METHODS To assess the impact of monetary policy and macroprudential policy on bank stability in Vietnam during the 2008-2018 period, the thesis uses quantitative in combination with quantitative research methods. First off, the author uses qualitative methods such as content analysis, statistical description, analysis and synthesis, induction, deduction, generalization, 14 and interview experts on the impact of each instrument in these policies and the interaction between these policies on bank stability in Vietnam. Then, research models are proposed. To estimate these models, a quantitative method with panel data regression is applied, including Pooled OLS, FEM, REM. In the case of existence of defects such as autocorrelation, heteroscedasticity, endogenous phenomenons, the estimation results of Pooled OLS, FEM, REM will be biased, so to overcome these defects, GMM method by Blundell and Bond (1998) is the most appropriate (Judson & Owen, 1999). The tests using in the model The author uses the tests related to the linear regression model: multi-collinear, autocorrelation, heteroskedasticity, endogenous. In addition, in the GMM estimation, the dissertation carries out some specific tests including Sargan test (Hansen test), Arellano – Bond (AR) test, Simultaneously, to ensure the robustness of instrument variables, it is required the number of groups is more than or equal to a number of instruments. 3.2. RESEARCH MODELS Models estimate the impact of monetary policy on bank stability Based on the research models of Altunbas, Gambacorta, and Marques-Ibanez (2010a), Altunbas, Gambacorta, and Marques-Ibanez (2010b), Altunbas, Gambacorta, and Marques- Ibanez (2012), Chen et al. (2017), de Moraes and de Mendonça (2019), Ngambou Djatche (2019), research models analyzing the impact of monetary policy on bank stability as following: Stabilityi,t = α0 + α1Stabilityi,t-1+ 𝛂j𝑴𝒐𝑷t + 𝜷𝒋𝑴𝑪t, + 𝜷𝒌𝑩𝑺𝑪i,t + 𝜺i,t (3.1) Models estimates impact of macroprudential policy on bank stability Based on the research models of Altunbas et al. (2018), Yến and Ngân (2016) on the impact of macroprudential policy on bank risks, and practicing on macroprudential policy 15 implementation, the research models assessing the impact of macroprudential policy on bank stability as following: Stabilityi,t = α0 + α1Stabilityi,t-1+ 𝛂j𝑴𝑷i,t + 𝜷𝒋𝑴𝑪t, + 𝜷𝒌𝑩𝑺𝑪i,t + 𝜺i,t (3.2) Models estimates the impact of monetary and macroprudential policies on bank stability Based on the research models of Bruno et al. (2017), Maddaloni and Peydró (2013) và Trung and Chung (2018), the research models as following: Stabilityi,t = α0 + α1Stabilityi,t-1+ 𝛂’j𝑴𝒐𝑷t + 𝛂j𝑴𝑷i,t + 𝜷𝒋𝑴𝑪t, + 𝜷𝒌𝑩𝑺𝑪i,t + 𝜺i,t (3.3) In addition, to assess the interaction between monetary and macroprudential policy on bank stability, after interviewing experts, the thesis uses the interaction variable is monetary policy × macroprudential policy, the research models as following: Stabilityi,t = α0 + α1Stabilityi,t-1+ 𝛂j𝑴𝒐𝑷𝒕×𝑴𝑷i,t + 𝜷𝒋𝑴𝑪t, + 𝜷𝒌𝑩𝑺𝑪i,t + 𝜺i,t (3.4) In these above models research: Stability: Dependent variables, are measured by lnZ-score and NPL (Non- performing loan to total loans). MoP: Independent variables, measure monetary policy variables. MP: Independent variables, measure macroprudential policy. MoP×MP: Interaction variables between monetary and macroprudential policies. MC: Control variables, describe the macro economies, including GDP, CPI. BSC: Control variables, describe the bank characteristics, including bank size, the cost operation to income operation ratio, and loan to total assets ratio. 3.3. VARIABLES IN RESEARCH MODELS 3.3.1. Bank stability The thesis uses the non-performing loan to total loans (NPL) and Z-score as two indicators representing bank stability. Abuzayed et al. (2018), Fernández et al. (2016), Jayakumar et al. (2018), Dwumfour (2017), Tuyền et al. (2017) were used these indicators in their researches. In particular, Z-score is calculated by the formula: 16 Z-scoreit = 𝑅𝑂𝐴𝑖𝑡+ 𝐸 𝐴𝑖𝑡 𝜎(𝑅𝑂𝐴)𝑖𝑡 (3.5) In which ROAit: is the return on total assets of bank i year t. E/Ait: is the ratio of equity to total assets of bank i year t. σ(ROA)it: is the standard deviation ROA of bank i year t. NPL is calculated by the formula: NPL = Debt of (group 3+group 4+group 5) Total loans (3.6) 3.3.2. Monetary policy indicators According to Cecchetti, Schoenholtz, and Fackler (2006), central banks can use several instruments to affect the money supply, interest rates, and some other indicators. Interest rates are one of these instruments. Chen et al. (2017) used a short-term interest rate. The choice of which interest rates depends on how to operate the monetary policy. Besides, interest rate, money supply is an indicator representing the intermediate objective of the monetary policy. Nguyen Thanh et al. (2017) uses a money base, rediscount interest rates, and reserve requirement when analyzing the impact of monetary policy on bank profits. Because in recent years, the SBV has rarely changed the reserve requirement, so the thesis uses the rediscount interest rate and money supply M2 to represent monetary policy. 3.3.3. Macroprudential policy indicators So far, no standard macroprudential policy instruments for all countries. The instruments’ countries depend on the level of economic-financial development, exchange rates, monetary policy. In Vietnam, based on circular no.36/2014/TT-NHNN date November 20, 2014, stipulating minimum safety limits and ratios for transactions performed by credit institutions and branches of foreign banks and expert opinions, the thesis uses 3 instruments to represent the macroprudential policy, including the capital adequacy ratio – CAR, the liquidity ratio (LIQ) and loan deposit ratio (LDR). These instruments are also consistent with Trung and Chung (2018), Yến and Ngân (2016). 17 The commercial banks have to maintain 9% for the capital adequacy ratio (CAR), 10% for the liquidity ratio and 80% of loan deposit ratio (according to circular no. 36/2014/TT- NHNN) 3.3.4. Interactive variable between monetary and macroprudential policy The thesis uses variable interaction between monetary and macroprudential policies is the lnM2×LDR variable. 3.3.5. Variables on bank characteristic: Bank size (BANKSIZE), cost-effective management (the cost operation to income operation ratio – CIR) and loan to total assets ratio (LOANTA) 3.3.6. Variables on macroeconomy: GDP growth and inflation rate - CPI 3.4. RESEARCH DATA The thesis used unbalanced panel data of 22 Joint Stock Commercial Banks during 2008-2018. All data was collected from their audited financial statements and annual reports. The M2 is collected from Asian Development Bank (ADB), rediscount interest rates were collect from regulations of the SBV in each period, then calculated the average for each year. The GDP and CPI were derived from the IMF database. After collecting data, the author calculated each variable then used software STATA 16 to test and regress and analyze research results. CHAPTER 4 RESEARCH RESULTS AND DISCUSSION 4.1. DESCRIPTIVE STATISTICS FOR RESEARCH DATA 4.2. DESCRIPTIVE STATISTICS FOR VARIABLES IN RESEARCH MODELS 4.2.1. Bank stability 4.2.2. Descriptive statistics 18 Table 4.3: Descriptive statistics Parameters Unit Obs Mean Std. Err Min Max Z-score 240 87.21843 187.0017 1.738938 2030.141 NPL % 237 0.022105 0.012669 0 0.089518 CAR % 236 0.139844 0.059118 0 0.4589 LIQ % 242 0.205854 0.135108 0.041585 0.994292 LDR % 242 0.873116 0.19209 0.235094 1.423927 M2 Bill VND 242 4860209 2409774 1622130 9211848 DIS % 242 0.063982 0.026053 0.04 0.118685 CIR % 242 0.886281 0.092792 0.40353 1.21875 BANKSIZE 242 18.2403 1.244142 14.69872 20.99561 LOANTA % 242 0.537874 0.136754 0.113904 0.851684 GDP % 242 0.061 0.005982 0.052 0.071 CPI % 242 0.079818 0.0667 0.006 0.231 Source: Author’s calculation from STATA 16 4.3. RESEARCH RESULTS 4.3.1. Results of the impact of monetary policy on bank stability Table 4.6. The results estimate the model of the impact of monetary policy on bank stability by SGMM method Variables Z-score dependent variable model NPL dependent variable model Coeff. Std. Err Coeff. Std. Err LnZ-score (t-1) .5985771*** .1358909 NPL(t-1) .4091006*** .061771 lnM2 -1.878998*** .5726109 .0113952*** .0015175 DIS -1.709974 5.704325 .1782749*** .0452754 CIR 4.165728*** 1.096949 -.0121511 .0091342 19 BANKSIZE 1.446562*** .4296254 -.0015921*** .0002584 LOANTA .443104 1.157995 -.0067875 .0099483 GDP 30.05392*** 10.31976 -.5112791*** .0638599 CPI -3.343678 2.832916 .0050515 .0253027 Cons -1.86869 2.416609 -.0981148 .0161118 The tests AR (1) p-value 0.005 0.028 AR (2) p-value 0.389 0.644 Hansen p-value 0.280 0.263 Number of groups 22 22 Number of instruments 21 22 F-test p-value 0.000 0.000 Obs 218 213 ***, **, * statistically significant at 1%, 5%, 10% Nguồn: Calculation results from STATA 16 software 4.3.2. Results of impact of macroprudential policy on bank stability Table 4.7. The results estimate the model of the impact of monetary policy on bank stability during the 2008-2018 period by SGMM method Variables Z-score dependent variable model NPL dependent variable model Coeff. Std. Err Coeff. Std. Err LnZ-score (t-1) .1085479 .1246443 NPL (t-1) .3037163* .1663397 CAR 11.69292*** 3.325072 .0320102 .0551049 LIQ 11.65443*** 4.056871 -.0591428*** .0177903 20 LDR 1.507406 3.122798 .0318361** .0318361 CIR 15.71469** 6.773425 .0318361 .014429 BANKSIZE .588054*** .1928372 .0008879 .0020477 LOANTA 2.074113 3.289676 -.0396737* .0210436 GDP 16.56165 14.10559 -.2128972* .1206882 CPI -4.623497 4.071213 .0254833 .0215086 Cons -28.24178 10.46021 -.0054342 .0331319 The tests AR (1) p-value 0.030 0.019 AR (2) p-value 0.153 0.491 Hansen p-value 0.692 0.947 Number of groups 22 22 Number of instruments 21 22 F-test p-value 0.000 0.000 Obs 215 208 ***, **, * statistically significant at 1%, 5%, 10% Nguồn: Calculation results from STATA 16 software 4.3.3. Results of the impact of monetary policy and macroprudential policy on bank stability Table 4.8. The results estimate the model of the impact of monetary policy and macroprudential policy on bank stability by SGMM method Variables LnZ-score dependent variable NPL depen