Tóm tắt Luận án Monetary policy transmission and bank lending channel in Vietnam

ployed to estimate the impact of monetary policy evolved overtime because of time- series econometrics development and changes in specific questions by theory. Tobin (1970) was the first economist who formally modeled the idea that the positive correlation between money and output could reflect the opposite view: the output might be caused by money. One of the first time-series econometric studies tries to estimate effects of monetary policy is Friedman and Meiselman (1963). They try to test the role of monetary policy in determining nominal income through equation: 𝑦"# = 𝑦" + 𝑝" = 𝑦(# + 𝑎*𝐴",-*.( + 𝑏*𝑚",-*.( + ℎ*𝑧",-*.( + 𝑢" (3.1) where: yn denotes the log of nominal income, equals to the sum of logs of output and the price level, A is a measure of autonomous expenditures, m is a monetary aggregate, and z is a vector of other variables relevant for explaining nominal income fluctuations. Because effects of macroeconomic variables are usually lagged, we suppose that estimated relationships between output and money are: 𝑦" = 𝑦( + 𝑎(𝑚" + 𝑎-𝑚",- + 𝑐-𝑧" + 𝑐5𝑧",- + 𝑢" (3.3) 3.1.3. Database in study of monetary policy transmission In studies of monetary policy transmission, data included macroeconomic elements such as policy rates, money supply, reserve requirement, GDP, unemployment, industrial production, exchange rate, and microeconomic elements that relate to specific transmission channels. 3.1.4. Proxy variables for monetary policy 3.1.4.1. Policy rates According to the traditional economic theory, central banks change money supply to influence interest rates rather than other economic variables. Nowadays, central bank mainly changes directly to policy rates to conduct monetary policy thus policy rates are often used as a good proxy for monetary policy. 3.1.4.2. Money supply Besides policy rates, money supply is one of the most important proxies of monetary policy. It is also the intermediate target that central bank aims in monetary policy conducting. 3.1.5. Variables of commercial bank characteristics in bank lending channel Economists agree that commercial bank characteristics have impacted on monetary policy transmission via BLC and they use three variables: scale, capital, and liquidity of banks (Kashyap and Stein, 2000). In which, Bank scale is logarithm of total asset, Bank liquidity is ratio of liquidity assets to total assets, in which total liquidity assets include total cash, deposits at central bank and other banks, investment securities, and short – term securities), Bank capital is ratio of bank capital to total assets. 3.2. Economic models for monetary policy transmission testing 3.2.1. VAR and related models 3.2.1.1. VAR model VAR was introduced by Sims (1972) and Sims (1980), it can be used by macro economists to quantify responses of macro variables that do not require strong conditions to determine shocks. Then VAR gradually becomes one of the most popular models used for time-series data over time. 3.2.1.2. SVAR model Besides VAR, structural VAR model (SVAR) is also popularly used in the research of monetary policy transmission in general and in particular with BLC. SVAR is widely used because SVAR helps clarify the structural shocks and unstructured shocks which standard VAR models can’t clarify, meanwhile SVAR also provides two tools to analyze monetary policy transmission: impulse response function and variance decomposition. 3.2.2. Cointegration models 3.2.2.1. ECM ECM is a dynamic system with characteristics that the deviation of current state from its long-run relationship will be fed into its short-run dynamics. ECM is a category of multiple time series models that directly estimates the speed at which a dependent variable returns to long-term equilibrium after a change in an independent variable (Engle and Granger, 1987). 3.2.2.2. VECM VECM is an innovation of ECM that adds error correction features to a multi-factor model such as a vector autoregression model. 3.2.2.3. ARDL If two or more series have cointegration but they are not stationary at the same level, we have to use ARDL. The ARDL of Pesaran and Shin (1998), Pesaran et al. (2001) has a number of features that many researchers give it some advantages over conventional cointegration testing. 3.2.3. DSGE model Many researchers study monetary policy transmission using VAR, but Dynamic stochastic general equilibrium modeling (DSGE) is also used in the case we have all the utility functions of economic agents such as households, firms, government, central bank (George et al., 2008, Cogley and Sargent, 2005). 3.2.4. GMM model for panel data GMM is often proposed to study monetary policy transmission through BLC in general and monetary policy transmission through BLC in the case of caring effects of bank characteristics on BLC. GMM helps to solve endogeneity and other problems in panel data (Arellano and Bond, 1991). In the GMM, widely alternatives in within estimation are methods such as Arellano-Bond differences in Arellano & Bond (1991) and Blundell-Bond system GMM (Blundell and Bond, 1998). GMM methods are considered superior to alternatives in handling endogeneity, heteroskedasticity, serial correction and identification (Hall, 2003). 3.3. Research methodologies for this study 3.3.1. Research procedures and testing hypothesizes This study goes into investigating the interest rate channel, asset price channel, exchange rate channel, and bank lending channel in Vietnam by using three main models: VAR, SVAR, and GMM in 4 steps as following: Step 1 and 2. In order to answer the study’s question 1, this study uses the results from literature review in chapter 2 to clarify the following testing hypothesizes: Hypothesis 1: the interest rate channel exists in Vietnam, Hypothesis 2: the exchange rate channel may or may not exist in Vietnam, Hypothesis 3: the asset price channel may or may not exist in Vietnam. In order to test these hypothesizes, this study builds VAR and SVAR to test three main monetary policy transmission channels in Vietnam including the interest rate channel, exchange rate channel, and asset price channel. Step 3 and 4. This study goes to measure the existence and the impact of commercial bank characteristics on bank lending channels by using the GMM model with the following hypothesizes: Hypothesis 4: the bank lending channel is existed in Vietnam, Hypothesis 5: the bank lending channel is affected by bank size, bank capital, bank liquidity and bank risk, Hypothesis 6: the BLC is affected by the 2008 global financial crisis. In order to test these hypothesizes, this study recruits the GMM model which has been used in many previous studies and is considered as an appropriate model for the bank lending channel with micro data from commercial banks. 3.3.2. Vietnam monetary policy transmission testing models 3.3.2.1. VAR model This study uses VAR model which is used in study of Aleem (2010). VAR (p) is formed: 𝜑*𝑌",*8*.( = 𝜃𝑋" + 𝜀" (3.18) in which: Yt is endogenous vector of domestic variables, Xt is exogenous variables, 𝜑, 𝜃 are coefficient vector, ɛ: residual vector. Exogenous variables include world commodity index (Compiworld), U.S policy rate (ius), and U.S. output (yus). Xt = [Compiworld ius yus] (3.19) Meanwhile, the endogenous vector expresses Vietnam economic factors which include Vietnam industrial production (IPVN) which presents for Vietnamese output, Vietnam price index (Prices), and Vietnam monetary policy rate (i) Yt = [IPVN Prices i] (3.20) In order to measure monetary policy transmission through the interest rate channel, exchange rate channel, and asset price channel, this study also incorporates endogenous variables which represent each specific channel including market interest rates, nominal effective exchange rate of USD/VND, and VNindex. 3.3.2.2. SVAR model This study uses the SVAR model from study of Neri and d'Italia (2004) then adjust to the form as Table 3.2. SVAR restriction matrix ɛcp 1 0 0 0 0 0 0 0 ucp ɛy a21 1 0 0 0 0 0 0 uy ɛms 0 a32 1 a34 a35 0 0 a38 ur ɛmd = 0 a42 a43 1 0 0 0 a48 um ɛexc a51 a52 a53 a54 1 0 0 0 uexc ɛle 0 a62 a63 a64 0 1 0 0 ulr ɛvni a71 a72 a73 a74 a75 a76 1 0 uvni ɛp a81 a82 a83 a84 a85 a86 a87 1 up Source: author’s building In which: cp (world commodity price index), y (Vietnam industrial production), r (Vietnam policy rate), ms (Vietnam money supply), md (Vietnam money demand), exc (Vietnam nominal exchange rate), le (Vietnam lending rate), vni (Vietnam stock index), and p (Vietnam consumer price index). Structural shocks include: ucp, uy, ur, um, uexc, ulr, uvni, and up are world price shock, domestic output shock, monetary policy shock, money demand shock, international shock (exchange rate shock), lending interest rate shock, asset price shock, and price shock respectively. 3.3.3. Bank lending channel testing model The model is used by Altunbas et al. (2010) has form: ∆ln (𝐿𝑜𝑎𝑛𝑠)*," = 𝛼∆ln (𝐿𝑜𝑎𝑛𝑠)*,",- + 𝛿H∆ln (𝐺𝐷𝑃)L,",H-H.( + 𝛽H∆iO,",H-H.( + ∅H∆iO,",H ∗ 𝐸𝐷𝐹*,",- + 𝜎H∆iO,",H ∗ 𝑆𝐼𝑍𝐸*,",--H.( + 𝛾H∆iO,",H ∗ 𝐿𝐼𝑄*,",- +-H.(-H.( 𝜋H∆iO,",H ∗ 𝐶𝐴𝑃*,",- + 𝜏𝑆𝐼𝑍𝐸*,",--H.( + 𝜗𝐿𝐼𝑄*,",- + 𝜖𝐶𝐴𝑃*,",- + 𝜑𝐿𝐿𝑃*,",- + 𝜔𝐸𝐷𝐹*,",- + 𝜀*," (3.33) Source: Altunbas et al. (2010). In which, Loan is outstanding loans of commercial banks, GDP is output, i is policy rate, EDF is expected default frequency, SIZE is bank size, LIQ is bank liquidity, CAP is bank capital, LLP is bank loan loss provision. The model equation has form: Δln(loan)i,t = αln(loan)i,t-1 + σ2ln(GDP)t-1 + ϕ1Δit + ϕ2Δit-1 + φ1Δit*SIZEi,t-1 + τ1Δit*LIQi,t-1 + Ω1Δit*LLPi,t- 1 + ¥1SIZEi,t-1 + ψ1Δit*CAPi,t-1 + ω1LIQi,t-1 + λ1CAPi,t-1 + µ1LLPi,t-1 + ɛi,t (3.34) Variables in this model are the same as Altunbas et al. (2010), loan growth is affected by monetary policy, in particular through interest rate tool. 3.3.4. Research data Firstly, the monthly data from 2003 to 2012 which are used in VAR and SVAR models including world commodity index (Compiworld), US policy rate (ius), and U.S. output (yus), Vietnam output (IPVN), Vietnam price index (Prices), and Vietnam monetary policy rate (i), market interest rate (LER), the nominal exchange rate of USD/VND (NEER), and VNindex (VNI) are compiled primarily from the International Monetary Fund (IMF), General statistics office of Vietnam, and State bank of Vietnam. We also collect the SBV’s rediscount rate (RDR), SBV’s refinance rate (RFR), and money supply (M2) in place of VNIBOR for Vietnams monetary policy rate in order to test the consistence of VAR and SVAR models. Meanwhile, the annual data which uses the GMM model are collected from financial reports of commercial banks in the 2003 – 2012 period, but some banks do not have full financial statements so unbalanced panel data is used, while the GDP and Vietnam policy rates including VNIBOR, refinance rate and rediscount rate are collected from ADB and SBV. 3.4. Summary This study uses VAR to test the existence of main transmission channels in Vietnam. Based on the results of VAR, this study is going to use SVAR with short-term restrictions to test again Vietnam monetary policy transmission through main channels in one system. Next, this study uses GMM with data from financial reports of commercial banks to test the existence of a bank lending channel. These research methodologies help us to ensure: the monetary policy transmission is tested by appropriate models and the estimation problems are solved to avoid mistakes in the conclusions. With models and variables that are defined in this section, the next chapter is going to use them for experimental studies. CHAPTER 4 EMPIRICAL EVIDENCES FROM VIETNAM 4.1. Monetary policy transmission 4.1.1. Data With macroeconomic data for VAR and SVAR models, this study collects and organizes a monthly time series. Statistical descriptions are detailed in table 4.1. Table 4.1. Data statistical description OIL LIBOR IPUS IPVN CPI VNIBOR LER NEER VNI M2 RDR RFR Mean 69.71 2.16 103.81 52,001. 10.26 8.68 12.76 17,409. 443.09 1014930. 6.28 8.16 Median 70.28 1.28 104.04 51,255. 8.41 7.51 11.18 16,115. 421.90 816798.0 4.90 6.60 Max 133.93 5.50 112.98 86,118. 28.36 17.57 20.25 21,013. 1,137. 2590871. 13.00 15.00 Min 28.13 0.25 91.70 25,122. 2.05 5.22 9.30 15,417. 136.20 163090. 3.00 4.80 Std. 24.81 1.90 4.80 16,939. 6.42 2.87 3.08 2,000. 233.89 726798. 3.23 3.15 N 120 120 120 120 120 120 120 120 120 120 120 120 Source: author’s calculation. After stationary processing, short-term relationships between variables are tested through Granger causality test (Granger, 1969, Granger, 1980). Table 4.3. Granger causality test results Variables H01: VNIBOR no Granger cause to variable H02: Variable no Granger cause to CPI F-Statistic p-value F-Statistic p-value DLER 9.398 0.000 0.470 0.758 DLNEER 2.778 0.030 1.044 0.388 DLVNI 1.135 0.344 0.185 0.946 H03: VNIBOR no Granger cause to CPI F-Statistic p-value 2.210 0.073 Source: author’s calculation. The Granger causality test is used for VNIBOR and CPI, and results show that VNIBOR has Granger causality to CPI at 10% significance level (H03). This demonstrates that monetary policy has impacted on price level but this impact can be transmitted through the interest rate channel (via DLER), exchange rate channel (via DLNEER), but not asset price channel (via DLVNI). 4.1.2. VAR model results VAR model is used to test the existence of three channels: the interest rate channel, exchange rate channel, and the asset price channel, one by one. Interest rate channel Figure 4.1. Impulse response function of CPI in IRC Source: Author’s calculation. Inflation reacts positively with Vietnams interbank offer rate and lending rates of commercial banks, which defines the existence of a cost channel in Vietnam. This result confirms the existence of IRC in Vietnam that also proves the hypothesis 1. Exchange rate channel Figure 4.2. Impulse response function for ERC Source: author’s calculation. -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to LIPVN -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to VNIBOR -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLER -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to CPI Response to Cholesky One S.D. Innovations ± 2 S.E. -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to LIPVN -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to VNIBOR -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLNEER -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to CPI Response to Cholesky One S.D. Innovations ± 2 S.E. Impulse response function results show no transmission of a monetary policy through the exchange rate channel since inflation almost has no response to exchange rate shocks. This result confirms the hypothesis 2 that ERC may be weak or almost not exist in Vietnam. Asset price channel Figure 4.3. Impulse response function of VAR(5) for APC Source: Author’s calculation. APC via stock price may not exist in Vietnam due to a low-developed stock market, while other financial intermediaries including commercial banks thrive and play an important role in capital markets. This result confirms the hypothesis 3 that APC is weak or not exist in Vietnam. 4.1.3. SVAR model results Figure 4.4. Impulse response function of SVAR -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to LIPVN -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to VNIBOR -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLVNI -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to CPI Response to Cholesky One S.D. Innovations ± 2 S.E. Source: author’s calculation. While responses of inflation to policy rate, money supply, and market interest rates reconfirm the existence of a cost channel in Vietnam the same as the results from previous VAR models. 4.1.4. Robustness check In order to check the robustness of the models, M2 of Vietnam is collected to replace VNIBOR in proxy for Vietnam’s monetary policy; IRC, ERC, and APC are tested by VAR model one by one. Figure 4.5. Impulse response function of VAR for IRC with DLM2 Source: author’s calculation. This result also confirms that policy rate is more effective in monetary policy conducting in comparing with money supply of SBV because the money supply hasn’t affected the lending rates of commercial banks. -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLOIL -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to LIPVN -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to VNIBOR -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLM2 -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLNEER -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLER -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLVNI -2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to CPI Response to Cholesky One S.D. Innovations ± 2 S.E. -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to LIPVN -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLM2 -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to DLER -1 0 1 2 3 1 2 3 4 5 6 7 8 9 10 11 12 Response of CPI to CPI Response to Cholesky One S.D. Innovations ± 2 S.E. When we use RDR to replace for VNIBOR, Impulse response function shows that CPI has a strong positive responses to DRDR, DLER so that IRC exists. The positive responses of CPI to policy rate and lending rate indicate that a cost channel is exists as previous results in VAR and SVAR models. The robustness test is continued with a new proxy for Vietnam monetary policy: refinance rate (RFR). Results show the same idea with the rediscount rate because SBV has changed both RDR and RFR have the same pattern. 4.2. Bank lending channel in Vietnam 4.2.1. Data We used a unique dataset of bank balance sheet items and macroeconomic variables for Vietnam over the period from 2003 to 2012. Our data was collected from different sources, bank characteristics data are hand collected from financial reports and annual reports, other data are from ADB and other sources. Table 4.9. GMM data description LnLOAN SIZE LIQ CAP LLP Δi LnGDP ΔRFR ΔRDR Mean 16.320 17.260 0.381 0.130 1.180 0.188 14.271 0.663 0.631 Maximum 19.990 20.242 0.823 0.712 7.425 6.143 14.795 5.330 -4.430 Minimum 11.478 12.198 0.036 0.027 0.032 -5.339 13.481 -3.920 5.750 Std. Dev. 1.700 1.556 0.149 0.103 1.005 3.847 0.403 2.800 3.094 N (ind.) 282 282 283 282 279 271 300 270 270 Source: author’s calculation. 4.2.2. GMM results and discussions In order to examine the bank lending channel, we should put all determinants into one model which is presented in table 4.11. Table 4.11. Bank lending channel with whole effects of bank characteristics Δln(loan)t Whole effects without risk Whole effects Coeff. P-Value Coeff. P-Value ln(loan)t-1 -0.595*** 0.000 -0.621*** 0.000 Ln(GDP) t-1 0.417*** 0.003 0.420*** 0.003 Δi -0.207*** 0.009 -0.212** 0.029 Δit-1 -0.021*** 0.002 -0.021*** 0.002 Δi*sizet-1 0.010** 0.019 0.010* 0.080 Δi*capt-1 0.041 0.449 0.044 0.439 Δi*liqt-1 0.012 0.700 0.013 0.721 Δi*llpt-1 0.039 0.979 sizet-1 0.172 0.205 0.197 0.155 capt-1 -0.093 0.773 -0.007 0.984 liqt-1 0.174 0.542 0.144 0.620 llpt-1 2.439 0.552 2.926 0.503 AR(-2) test (p-value) 0.955 0.913 Sargan test (p-value) 0.128 0.125 N 182 182 *, **, *** indicate significance level of 10%, 5% and 1% respectively Source: author’s calculation. The results are consistent with all above findings with these results means that a bank lending channel exists in Vietnam, and especially Vietnamese commercial banks have risk-taking action in responding with monetary policy. But as the study of Nguyen and Boateng (2013), then Nguyen and Boateng (2015a) and Nguyen and Boateng (2015b) suggest that when we use commercial bank characteristics to investigate the bank lending channel we will face to problem of data normalization due to the heteroskedasticity between commercial banks thus we should standardize commercial bank data to be more reliable measurement of bank characteristics. Thus, this study standardizes the data of commercial bank characteristics by counting the mean of each characteristics including bank size, bank capital, bank liquidity and bank loan loss provision for each bank, then we minus the value of each characteristic for its mean one by one for each bank. Table 4.13 presents estimation results when we put all variables into one model, these results are interesting. Table 4.13. Bank lending channel with whole effects by new measure of bank characteristics Δln(loan)t Whole effects without risk Whole effects Coeff. P-Value Coeff. P-Value ln(loan)t-1 -0.548*** 0.000 -0.548*** 0.000 Ln(GDP) t-1 0.323*** 0.005 0.322*** 0.005 Δi -0.040*** 0.000 -0.039*** 0.000 Δit-1 -0.025*** 0.000 -0.025*** 0.000 Δi*sizet-1 0.043*** 0.000 0.042*** 0.000 Δi*capt-1 0.222*** 0.001 0.217*** 0.001 Δi*liqt-1 -0.006 0.901 0.001 0.979 Δi*llpt-1 0.773 0.593 sizet-1 0.188 0.101 0.189* 0.099 capt-1 0.470 0.111 0.478 0.105 liqt-1 0.323 0.186 0.324 0.185 llpt-1 1.960 0.575 1.474 0.683 AR(-2) test (p-value) 0.967 0.955 Sargan test (p-value) 0.305 0.275 N 182 182 *, **, *** indicate significance level of 10%, 5% and 1% respectively Source: author’s calculation. These results also confirm the consistent of our findings, this also indicates the significant effects of bank capital on bank lending when Vietnamese commercial banks respond to monetary policy rate. 4.2.3. Robustness test Firstly, rediscount rate (RFR) and refinance rate (RDR) are used to replace for VNIBOR in proxy for Vietnamese monetary policy, but all the bank characteristics are measured by standardized way as we presented in previous section due to better results in GMM. Estimation results from GMM with RFR and RDR indicate the same evidence as in GMM with VNIBOR in all aspects. Continue with the GMM model, this study replaces VNIBOR by rediscount rate (RDR). Results show the same evidences as in the GMM model with RFR and VNIBOR. So that, we can confirm the consistency of our findings in this section. 4.2.4. The effects of the crisis on the bank lending channel in Vietnam In order to investigate the effects of the 2008 global financial crisis on the bank lending channel in Vietnam, we can divide the data into two sub-periods including pre-crisis and post-crisis periods, but due to the lack of data this study does not use this method. Instead, we use the proxy for the 2008 global financial crisis, this is VIX (the implied volatility of S&P 500 options contracts) that is suggested to be used to investigate the impacts of the crisis (Druck et al., 2015). The results show the same evidence as previous estimations in section 4.2.1 and 4.2.2 for the same variables. While the significant positive effects of VIX on bank lending indicate that Vietnamese commercial banks expand their lending in crisis periods, that means the bank lending channel is stronger in crisis periods. Meanwhile, the interaction terms between policy rate, bank size, bank capital and VIX have significant positive effects on bank lending of the Vietnamese commercial banks and indicate that the bank lending channel is stronger in larger banks and higher capitalized bank in crisis. These results confirm the hypothesis 6. Table 4.18. The impacts of the 2008 global financial crisis on bank lending channel in Vietnam Δln(loan)t Bank lending channel with VNIBOR Bank lending channel with RDR Bank lending channel with RFR Coeff. P-Value Coeff. P-Value Coeff. P-Value ln(loan)t-1 -0.536*** 0.000 -0.671*** 0.000 -0.672*** 0.000 Ln(GDP) t-1 0.372*** 0.003 0.474*** 0.000 0.476*** 0.000 Δi -0.041*** 0.000 -0.043*** 0.000 -0.048*** 0.000 Δit-1 -0.031*** 0.000 -0.030*** 0.000 -0.032*** 0.000 Δi*sizet-1*vix 0.002*** 0.000 0.001*** 0.000 0.002*** 0.000 Δi*capt-1*vix 0.006*** 0.007 0.006** 0.031 0.006** 0.031 Δi*liqt-1*vix 0.000 0.917 -0.001 0.604 -0.001 0.603 Δi*llpt-1*vix 0.024 0.650 0.036 0.563 0.041 0.556 sizet-1 0.116 0.324 0.233** 0.046 0.230** 0.049 capt-1 0.060 0.843 0.540* 0.067 0.509* 0.084 liqt-1 0.354 0.152 0.364 0.153 0.363 0.154 llpt-1 2.634 0.467 2.493 0.495 2.418 0.507 Vix 0.010*** 0.001 0.007*** 0.011 0.007*** 0.010 AR(-2) test (p-value) 0.983 0.804 0.791 Sargan test (p-value) 0.357 0.182 0.180 N 182 182 182 *, **, *** indicates significance level of 10%, 5% and 1% respectively. Source: author’s calculation. CHAPTER 5 CONCLUSIONS AND POLICY IMPLICATIONS 5.1. Introduction This concluding chapter summarizes the empirical findings of this thesis. 5.2. Review of research questions, methodology and findings 5.2.1. Research question 1: “Does an interest rate channel, exchange rate channel and asset price channel exist in Vietnam?” Results of VAR and SVAR models indicated the first main findings of this thesis: the cost channel existed in Vietnam, while they can not find the evidences of the exchange rate channel and asset price channel that means they may be weak or non-existent in Vietnam. The evidence of cost channel puts Vietnamese policy makers into a tough dilemma situation as monetary policy is not effective in controlling inflation. 5.2.2. Research question 2: Does a bank lending channel exist in Vietnam? And does bank size, bank capital, bank liquidity and bank risk have an effect and the 2008 global financial crisis effect on bank lending channel in Vietnam? By using the GMM model, the results have some main findings: Firstly, the loan level of Vietnamese commercial banks is converged which means that small Vietnamese commercial banks grow their credit portfolio to the level of large banks which will increase the competition in banking system. Secondly, the monetary policy has a strong effect on bank lending in Vietnam including the current effects and lag effects on the following year, this means that the monetary policy transmits through the credit channel in Vietnam. Thirdly, the main findings and the main aim of this study is to examine the existence of a bank lending channel. Fourthly, bank size has a significant positive effect on bank lending of Vietnamese commercial banks, the bank capital also has a significant positive effect on bank lending in responding to monetary policy. Fifthly, bank lending channel is stronger in crisis, besides that bank size and bank capital have more impacts on bank lending channel in crisis. 5.3. Academic contributions This thesis makes some important contributions both to the literature on the monetary policy transmission and the bank lending channel. The results for research question 1 in Chapter 4 provide strong statistical evidence of a cost channel, while the results in Chapter 4 also can not find the statistical evidence of an exchange rate channel and an asset price channel in Vietnam. Thus, as far as it could be ascertained, this is the first study with putting all literature together to test all channels of monetary policy in one model for a transition economy as Vietnam. With regard to methodologies in testing the monetary policy transmission, many studies which presented in section 2.2 in Chapter 2 show that they usually study each channel without consider all channels in one system. This result shows that policy rates are more effective for measuring monetary policy than money supply, especially in small open economy such as Vietnam. This thesis also contributes to the literature in terms of a bank lending channel in that the bank size, bank capital are more important for an emerging market as in Vietnam. Therefore, along with bank size and bank capital, the results for controlled variables related to research question 2, such as the lag of a loan level and economic growth are important. This adds to the credit channel literature that both credit demand and credit supplies are important in monetary policy transmission through a credit channel. In terms of methodology, this thesis also offers a number of enhancements on existing studies. It examines the results (related to research question 2) with a variety of measurement methods for bank characteristics which include several approaches that account for endogeneity, and the results remain very robust with such estimations. 5.4. Policy implications 5.4.1. Choosing monetary policy tools In Vietnam, SBV can use open market operation, policy rate (base rate, rediscount rate and refinance rate), reserve requirement, and other tools as stated in the 2010 state bank act. As investigation, both rediscount rate and refinance rate are effective in monetary policy transmission so that SBV can use both rates. So that, it is better for SBV if they focus more on the rediscount rate that is better in identifying market expectations and navigating market interest rates. In addition, if SBV uses the rediscount rate, they can limit the risk-taking activities of commercial banks in credit activities that help stabilize the banking system for Vietnam. 5.4.2. Appling unconventional monetary policies In normality, a central bank can conduct monetary policy easier than some tough cases such as deflation, near nil interest rates, liquidity traps, or crisis. Thus, quantitative easing program and inflation targeting policies are launched in many countries to tackle these dilemmas. Vietnam is going to integrate more into the international market that will put more problems on Vietnam’s authorities, especially SBV in monetary policy conducting, thus SBV should learn new innovations in monetary policy conducting from other countries to apply to Vietnam in the future. 5.4.3. Developing debt and equity markets As stated, Vietnam has to develop financial markets including markets for debts and equities more and more. Despite the development of the stock market in Vietnam with the initiations of HSX and HNX in 2000 and 2006, the number of enterprises listed on the stock market is small. Thus, Vietnam’s enterprises are limited in sources for funds, they still heavily rely on the banking system. So Vietnam needs to develop capital markets which focus on debt markets to supply funds for the economy. 5.4.4. Capability of commercial banks Besides the changes in the monetary policy and other financial markets, Vietnam has to make improvements in the banking system to make it safer and more efficient. Firstly, Vietnamese commercial banks need to raise capital to ensure flexibility and defensive capabilities with macroeconomic shocks. Secondly, because the size and liquidity of commercial banks have a significant impact on the monetary policy transmission and they respond strongly to monetary policy shocks, thus Vietnamese commercial banks should: consider the overall macro-economic factors, credit demand, their size and liquidity to plan an appropriate credit growth plan in business planning. 5.4.5. Risk of commercial banks Liquidity risk was one of the biggest problems in the banking system in the instability stage from 2008 to 2012, which has reduced the financial health of Vietnamese commercial banks. SBV should divide commercial banks into some small groups according to criteria: capital, asset liquidity, and credit risk, then they use suitable tools for monetary policy conducting. Regarding the credit policy, SBV should eliminate limitation regulations in credit growth for all commercial banks, they should set a credit growth rate for each bank group that depends on capital, size, liquidity and risk. 5.5. Limitations and suggestions for further research Besides contributions, this study has some limitations. Although, they are not significant to overall meanings of study but they need to be consider: Firstly, this study has investigated three main channels in monetary policy transmission, but it does not include the expectation channel, and some sub-channels in the credit channel such as balance sheet channel, cash flow channel, unexpected price level channel, and household effects liquidity. Secondly, the study about the bank lending channel has not incorporated the macroeconomic determinants and banking sector determinants into models to test the effects of these factors on the bank lending channel. Thirdly, this study was done using data from the 2003 – 2012 periods./. References ALEEM, A. 2010. Transmission mechanism of monetary policy in India. Journal of Asian Economics, 21, 186-197. ALTUNBAS, Y., FAZYLOV, O. & MOLYNEUX, P. 2002. Evidence on the bank lending channel in Europe. Journal of Banking & Finance, 26, 2093-2110. ALTUNBAS, Y., GAMBACORTA, L. & MARQUES-IBANEZ, D. 2009. Securitisation and the bank lending channel. European Economic Review, 53, 996-1009. ALTUNBAS, Y., GAMBACORTA, L. & MARQUES-IBANEZ, D. 2010. Bank risk and monetary policy. Journal of Financial Stability, 6, 121-129. ALTUNBAS, Y., GAMBACORTA, L. & MARQUES-IBANEZ, D. 2012. Do bank characteristics influence the effect of monetary policy on bank risk? Economics Letters, 117, 220-222. ANGELONI, I. & EHRMANN, M. 2003. Monetary transmission in the euro area: early evidence. Economic policy, 18, 469-501. APERGIS, N. & CHRISTOU, C. 2015. The behaviour of the bank lending channel when interest rates approach the zero lower bound: Evidence from quantile regressions. Economic Modelling, 49, 296- 307. ARAÚJO, E. 2015. Monetary policy objectives and Money’s role in U.S. business cycles. Journal of Macroeconomics, 45, 85-107. ARELLANO, M. & BOND, S. 1991. Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The review of economic studies, 58, 277-297. ASAKO, K., KUNINORI, M., INOUE, T. & MURASE, H. 1989. Tochi Hyouka to Tobin no q/Multiple q no Keisoku (Land Valuation and Tobin’s q: Estimation by Multiple q). Japan Development Bank, Keizai Keiei Kenkyu, 10-3. BARRAN, F., COUDERT, V. & MOJON, B. 1996. The transmission of monetary policy in the European countries, Citeseer. BERNANKE, B. S. Alternative explanations of the money-income correlation. Carnegie-rochester conference series on public policy, 1986. Elsevier, 49-99. BERNANKE, B. S. & BLINDER, A. S. 1988. Credit, Money, and Aggregate Demand. The American Economic Review, 435-439. BERNANKE, B. S. & BLINDER, A. S. 1992. The federal funds rate and the channels of monetary transmission. The American Economic Review, 901-921. BERNANKE, B. S. & GERTLER, M. 1995. Inside the black box: the credit channel of monetary policy transmission. National bureau of economic research. BERNANKE, B. S., LOWN, C. S. & FRIEDMAN, B. M. 1991. The credit crunch. Brookings papers on economic activity, 205-247. BLINDER, A. S., EHRMANN, M., FRATZSCHER, M., DE HAAN, J. & JANSEN, D.-J. 2008. Central bank communication and monetary policy: A survey of theory and evidence. National Bureau of Economic Research. BLUNDELL, R. & BOND, S. 1998. Initial Conditions and Moment Restrictions in Dynamic Panel Data Models. Journal of Econometrics, 87, 115-143. BREDIN, D., FITZPATRICK, T. & PATRICK, O. R. 2001. Retail Interest Rate Pass-Through: The Irish Experience. Central Bank of Ireland. BURGSTALLER, J. 2003. Interest Rate Transmission to Commercial Credit Rates in Austria. Department of Economics, Johannes Kepler University Linz, Austria. BUTI, M., ROEGER, W. & IN'T VELD, J. Monetary and fiscal policy interactions under a Stability Pact. t Veld, Jan, Monetary and Fiscal Policy Interactions Under a Stability Pact (February 1, 2001). Fiscal Rules Conference, 2001a. 279. BUTI, M., ROEGER, W. & VELD, J. I. T. 2001b. Stabilizing Output and Inflation: Policy Conflicts and Co-operation under a Stability Pact. JCMS: Journal of Common Market Studies, 39, 801-828. CANTERO-SAIZ, M., SANFILIPPO-AZOFRA, S., TORRE-OLMO, B. & LÓPEZ-GUTIÉRREZ, C. 2014. Sovereign risk and the bank lending channel in Europe. Journal of International Money and Finance, 47, 1-20. CECCHETTI, S. G. 1995. Distinguishing theories of the monetary transmission mechanism. Federal Reserve Bank of St. Louis working paper. CECCHETTI, S. G. & KRAUSE, S. 2002. Central bank structure, policy efficiency, and macroeconomic performance: exploring empirical relationships. Federal Reserve Bank of St. Louis Review, 84. CHRISTIANO, L. J., EICHENBAUM, M. & EVANS, C. L. 1999. Monetary policy shocks: What have we learned and to what end? Handbook of macroeconomics, 1, 65-148. CHRYSTAL, A. & MIZEN, P. 2002. Modelling credit in the transmission mechanism of the United Kingdom. Journal of Banking & Finance, 26, 2131-2154. CICCARELLI, M., MADDALONI, A. & PEYDRÓ, J.-L. 2014. Trusting the bankers: A new look at the credit channel of monetary policy. Review of Economic Dynamics. COGLEY, T. & SARGENT, T. J. 2005. Drifts and volatilities: monetary policies and outcomes in the post WWII US. Review of Economic dynamics, 8, 262-302. DABLA-NORRIS, E. & FLOERKEMEIER, H. 2006. Transmission mechanisms of monetary policy in Armenia: evidence from VAR analysis. International Monetary Fund. DE BONDT, G. 2002. Retail bank interest rate pass-through: new evidence at the Euro area level. European Central Bank. DE BONDT, G. 2004. The balance sheet channel of monetary policy: First empirical evidence for the euro area corporate bond market. International Journal of Finance & Economics, 9, 219-228. DEMIR, İ. 2014. Monetary policy responses to the exchange rate: Empirical evidence from the ECB. Economic Modelling, 39, 63-70. DIEPPE, A. & MCADAM, P. 2006. Monetary policy under a liquidity trap: Simulation evidence for the euro area. Journal of the Japanese and International Economies, 20, 338-363. DRAKOS, A. A. & KOURETAS, G. P. 2015. The conduct of monetary policy in the Eurozone before and after the financial crisis. Economic Modelling, 48, 83-92. DRUCK, P., MAGUD, N. E. & MARISCAL, R. 2015. Collateral Damage: Dollar Strength and Emerging Markets’ Growth. IMF working paper. IMF. EHRMANN, M. 2004. Firm size and monetary policy transmission: evidence from German business survey data. CESifo working papers. EHRMANN, M. & WORMS, A. 2004. Bank networks and monetary policy transmission. Journal of the European Economic Association, 2, 1148-1171. ENGLE, R. F. & GRANGER, C. W. 1987. Co-integration and error correction: representation, estimation, and testing. Econometrica: journal of the Econometric Society, 251-276. FERRI, G. & KANG, T. S. 1999. The credit channel at work: lessons from the financial crisis in Korea. Economic Notes, 28, 195-221. FRIEDMAN, B. M. 1956. The Quantity Theory of Money: A Restatement. Studies in the Quantity Theory of Money, Chicago University Press, Chicago. FRIEDMAN, B. M. 1999. The future of monetary policy: the central bank as an army with only a signal corps? International Finance, 2, 321-338. FRIEDMAN, M. & MEISELMAN, D. 1963. The Relative Stability of the Investment Multiplier and Monetary Velocity in the United States, 1897-1958. Stabilization Policies, Englewood Cliffs. New Jersey: Prentice-Hall. GEORGE, E. I., SUN, D. & NI, S. 2008. Bayesian stochastic search for VAR model restrictions. Journal of Econometrics, 142, 553-580. GERAATS, P. M. 2002. Central Bank Transparency. The economic journal, 112, F532-F565. GIRARDIN, E. & MOUSSA, Z. 2011. Quantitative easing works: Lessons from the unique experience in Japan 2001–2006. Journal of International Financial Markets, Institutions and Money, 21, 461- 495. GOLINELLI, R. & ROVELLI, R. 2005. Monetary policy transmission, interest rate rules and inflation targeting in three transition countries. Journal of Banking & Finance, 29, 183-201. GOODHART, C. & HOFMANN, B. Asset prices, financial conditions, and the transmission of monetary policy. conference on Asset Prices, Exchange Rates, and Monetary Policy, Stanford University, 2001. 2-3. GRANGER, C. W. 1969. Investigating causal relations by econometric models and cross-spectral methods. Econometrica: Journal of the Econometric Society, 424-438. GRANGER, C. W. 1980. Testing for causality: a personal viewpoint. Journal of Economic Dynamics and control, 2, 329-352. HALL, A. R. 2003. Generalized method of moments. A Companion to Theoretical Econometrics, 230-255. HANNAN, T. H. & LIANG, J. N. 1993. Inferring market power from time-series data: The case of the banking firm. International Journal of Industrial Organization, 11, 205-218. HAYASHI, F. & INOUE, T. 1991. The relation between firm growth and Q with multiple capital goods: theory and evidence from panel data on Japanese firms. Econometrica, 59, 731-753. HOLTEMÖLLER, O. 2003. Money Stock, Monetary Base and Bank Behavior in Germany, Geldmenge, Geldbasis und Bankenverhalten in Deutschland. Journal of Economics and Statistics (Jahrbuecher fuer Nationaloekonomie und Statistik), 223, 257-278. HOSHI, T. & KASHYAP, A. K. 1990. Evidence on q and investment for Japanese firms. Journal of the Japanese and International Economies, 4, 371-400. HÜLSEWIG, O., MAYER, E. & WOLLMERSHÄUSER, T. 2009. Bank behavior, incomplete interest rate pass-through, and the cost channel of monetary policy transmission. Economic Modelling, 26, 1310-1327. HULSEWIG, O., WINKER, P. & WORMS, A. 2001. Bank Lending and the Transmission of Monetary Policy: A VECM Analysis for Germany. International University in Germany Working Paper. 529 ed. Germany: International University. IACOVIELLO, M. & MINETTI, R. 2008. The credit channel of monetary policy: Evidence from the housing market. Journal of Macroeconomics, 30, 69-96. ISSING, O. 2004. The role of macroeconomic projections within the monetary policy strategy of the ECB. Economic Modelling, 21, 723-734. JEAN LOUIS, R. & BALLI, F. 2013. Low-inflation-targeting monetary policy and differential unemployment rate: Is monetary policy to be blamed for the financial crisis?—Evidence from major OECD countries. Economic Modelling, 30, 546-564. KAKES, J. & STURM, J.-E. 2002. Monetary policy and bank lending: Evidence from German banking groups. Journal of Banking & Finance, 26, 2077-2092. KASHYAP, A. K. & STEIN, J. C. The impact of monetary policy on bank balance sheets. Carnegie- Rochester Conference Series on Public Policy, 1995. Elsevier, 151-195. KASHYAP, A. K. & STEIN, J. C. 2000. What do a million observations on banks say about the transmission of monetary policy? American Economic Review, 407-428. KASHYAP, A. K., STEIN, J. C. & WILCOX, D. W. 1993. Monetary policy and credit conditions: Evidence from the composition of external finance. American Economic Review, 83, 78-98. KIM, H. E. 1999. Was the credit channel a key monetary transmission mechanism following the recent financial crisis in the Republic of Korea?, World Bank Publications. KISHAN, R. P. & OPIELA, T. P. 2000. Bank size, bank capital, and the bank lending channel. Journal of Money, Credit and Banking, 121-141. KLEIMEIER, S. & SANDER, H. 2000. Regionalisation versus globalisation in European financial market integration: Evidence from co-integration analyses. Journal of banking & finance, 24, 1005-1043. LÂN, C. K. 2012. Nghiên cứu thực nghiệm về truyền dẫn chính sách tiền tệ qua kênh tín dụng tại Việt Nam. Tạp chí Ngân hàng, 13. LEITH, C., MOLDOVAN, I. & ROSSI, R. 2015. Monetary and fiscal policy under deep habits. Journal of Economic Dynamics and Control, 52, 55-74. LEROY, A. 2014. Competition and the bank lending channel in Eurozone. Journal of International Financial Markets, Institutions and Money, 31, 296-314. LIU, C. & MINFORD, P. 2014. How important is the credit channel? An empirical study of the US banking crisis. Journal of Banking & Finance, 41, 119-134. LUCAS, R. E. 1980. Two illustrations of the quantity theory of money. The American Economic Review, 1005-1014. LYONNET, V. & WERNER, R. 2012. Lessons from the Bank of England on ‘quantitative easing’and other ‘unconventional’monetary policies. International Review of Financial Analysis, 25, 94-105. MANKIW, N. G. & TAYLOR, M. P. 2011. Macroeconomics, South-Western, Cengage Learning. MELITZ, J. 2000. Some Cross-country Evidence about Fiscal Policy Behaviour and Consequences for EMU, Department of Economics, Fraser of Allander Institute. MISHRA, P. & MONTIEL, P. J. 2012. How Effective Is Monetary Transmission in Low-Income Countries? A Survey of the Empirical Evidence. International Monetary Fund. MOJON, B. 2001. Financial structure and the interest rate channel of ECB monetary policy. Economie et Prevision, 147, 89-115. MOJON, B. T., SMETS, F. & VERMEULEN, P. 2002. Investment and monetary policy in the euro area. Journal of Banking & Finance, 26, 2111-2129. MUKHERJEE, S. & BHATTACHARYA, R. 2011. Inflation targeting and monetary policy transmission mechanisms in emerging market economies. International Monetary Fund. NERI, S. & D'ITALIA, B. 2004. Monetary policy and stock prices: theory and evidence, Banca d'Italia. NGUYEN, V. H. T. & BOATENG, A. 2013. The impact of excess reserves beyond precautionary levels on Bank Lending Channels in China. Journal of International Financial Markets, Institutions and Money, 26, 358-377. NGUYEN, V. H. T. & BOATENG, A. 2015a. An analysis of involuntary excess reserves, monetary policy and risk-taking behaviour of Chinese Banks. International Review of Financial Analysis, 37, 63- 72. NGUYEN, V. H. T. & BOATENG, A. 2015b. Bank excess reserves in emerging economies: A critical review and research agenda. International Review of Financial Analysis, 39, 158-166. OGAWA, K. & SUZUKI, K. 1998. Land value and corporate investment: evidence from Japanese panel data. Journal of the Japanese and International Economies, 12, 232-249. ORLOWSKI, L. T. 2015. Monetary expansion and bank credit: A lack of spark. Journal of Policy Modeling, 37, 510-520. PERERA, A., RALSTON, D. & WICKRAMANAYAKE, J. 2014. Impact of off-balance sheet banking on the bank lending channel of monetary transmission: Evidence from South Asia. Journal of International Financial Markets, Institutions and Money, 29, 195-216. PESARAN, M. H. & SHIN, Y. 1998. An autoregressive distributed-lag modelling approach to cointegration analysis. Econometric Society Monographs, 31, 371-413. PESARAN, M. H., SHIN, Y. & SMITH, R. J. 2001. Bounds testing approaches to the analysis of level relationships. Journal of applied econometrics, 16, 289-326. PETTURSON, T. 2001. The transmission mechanism of monetary policy: Analyzing the financial market pass-through. Working paper. RAMEY, V. How important is the credit channel in the transmission of monetary policy? Carnegie- Rochester Conference Series on Public Policy, 1993. Elsevier, 1-45. ROLNICK, A. J. & WEBER, W. E. 1994. Inflation, money, and output under alternative monetary standards. Federal Reserve Bank of Minneapolis. SÁNCHEZ, M. 2012. Inflation uncertainty and unemployment uncertainty: Why transparency about monetary policy targets matters. Economics Letters, 117, 119-122. SANDER, H. & KLEIMEIER, S. 2002. Asymmetric adjustment of commercial bank interest rates in the Euro area: An emperical investigation into interest rate pass-through. Kredit und Kapital, 2, 161- 192. SANDER, H. & KLEIMEIER, S. 2004. Convergence in euro-zone retail banking? What interest rate pass- through tells us about monetary policy transmission, competition and integration. Journal of International Money and Finance, 23, 461-492. SIMS, C. A. 1972. Money, income, and causality. The American Economic Review, 540-552. SIMS, C. A. 1980. Macroeconomics and reality. Econometrica: Journal of the Econometric Society, 1-48. SPYROMITROS, E. & TUYSUZ, S. 2012. Do monetary policy transparency, independence and credibility enhance macro-financial stability? International Journal of Economics and Finance, 4, p44. SVENSSON, L. E. 2000. Open-economy inflation targeting. Journal of international economics, 50, 155- 183. TAYLOR, J. B. 1995. The monetary transmission mechanism: an empirical framework. The Journal of Economic Perspectives, 11-26. THÀNH, S. Đ. & HẰNG, V. T. M. 2008. Nhập môn tài chính tiền tệ (1 ed.). , Hồ Chí Minh: NXB Đại học quốc gia Tp.HCM. TOBIN, J. 1970. Money and income: post hoc ergo propter hoc? The Quarterly Journal of Economics, 301- 317. TOOLSEMA, L. A., STURM, J.-E. & DE HAAN, J. 2001. Convergence of Monetary Transmission in EMU New Evidence. CESifo Group Munich. VAN DER CRUIJSEN, C. & DEMERTZIS, M. 2007. The impact of central bank transparency on inflation expectations. European Journal of Political Economy, 23, 51-66. VON KALCKREUTH, U. 2001. Monetary transmission in Germany: New perspectives on financial constraints and investment spending. Discussion paper Series 1/Volkswirtschaftliches Forschungszentrum der Deutschen Bundesbank. WORMS, A. 2001. The reaction of bank lending to monetary policy measures in Germany.