Bank lending channel and monetary policy transmission in Vietnam

when the latter effect depends on the net asset position of the consumer. Third, in the case of a floating exchange rate regime, interest rate fluctuations will cause the exchange rate to fluctuate, thereby changing price competitive advantage and affecting net exports. Interest rate fluctuations will also have an impact on the supply side of the economy through the substitution of labor supply across countries.

The interest channel theory assumes that financial intermediaries do not have any special role in the economy. Bernanke and Blinder (1988) show that the traditional interest rate channel is based on at least one of three assumptions: 1) loans and bonds are fully interchangeable with respect to borrowers, and 2) the Loans and bonds are completely interchangeable for lenders, or 3) demand for goods is not sensitive to lending rates .

The starting point of the credit view is the rejection of the assumption that bonds and bank loans are fully interchangeable. Bernanke and Gertler (1995) state that: “according to the credit channel theory, the direct effect of monetary policyThe rise in interest rates is diffused through a change in the external finance premium, the external capital reward being the difference in cost between the externally raised capital (from the issuance of bonds, stocks, bank loans) and capital mobilized from within (from retained earnings). This degree of disparity reflects credit market imperfections (owner and agent problems; appraisal, monitoring, and recovery costs; asymmetric information problem, causing selection) adversarial and moral hazard), which creates the difference between the lender's expected return and the borrower's costs. From a credit point of view, a change in monetary policy that raises or lowers market interest rates will tend to shift the external capital reward in the same direction.

Based on the assumption of information imperfections in financial markets, credit channels assume an active role in the provision of bank credits. Thus, the credit channel stipulates that monetary policy not only affects

credit demand, but also can affect credit supply. Bernanke and Gertler (1995) define two “subchannels” of the credit channel:

The balance sheet channel (also known as the broad credit channel) with a focus on the potential impact on a borrower's balance sheet and income statement.

- Bank lending channel (also known as narrow credit channel) with a focus on providing loans by credit institutions.

The balance sheet channel does not focus specifically on bank credit, but is generally more concerned with the overall capital supply. This channel can work even if loans and bonds are completely interchangeable in the balance sheets of banks and businesses. The balance sheet channel is based on the assumption that the external funding reward will depend on the borrower's financial position.

Changes in monetary policy can affect the risk profile of borrowers, as higher interest rates are expected to weaken the borrower's financial position. According to Bernanke and Gertler (1995), a tight monetary policy will directly weaken the balance sheet of borrowers in two ways. First, higher interest rates will increase interest costs, weakening the borrower's financial position. Second, an increase in interest rates is often associated with a decrease in property prices, which reduces the value of a borrower's collateral. Furthermore, a tight monetary policy can affect the financial position of borrowers (enterprises) indirectly, by reducing the firm's demand for output, while Other fixed costs are not adjusted in the short run.

Maybe you are interested!

The bank lending channel focuses more closely on the special role of banks in the monetary transmission mechanism, since monetary policy can also affect the

external funding by changing the supply of bank credit. This channel can only work under the following conditions:

- Monetary policy must have a significant impact on the supply (or interest rate) of bank credit. The model designed by Bernanke and Blinder (1988) shows that a tight monetary policy will affect the reserve position of banks, affecting the supply of bank credit. The key assumption is that, after a fall in reserves, banks are not able to widely shift to raising capital in the bond market because of the cost differential between external funding and funding. and funds raised from within (rewards from external sources). In other words, deposits and bonds are not fully interchangeable for banks.

Bank credit and bonds are not fully interchangeable for borrowers, and changes in the supply of credit must affect real activity in the economy. Bernanke and Gertler (1995) argue that this condition is very reasonable since banks are still the main source of credit, specializing in overcoming information problems and other barriers in credit markets.

Balance sheet channels and bank lending channels should not be viewed as an alternative to traditional monetary transmission. Bernanke and Gertler (1995) emphasize that: “we do not think of the credit channel as a separate channel that can replace the traditional monetary transmission mechanism, but rather as a set of amplifying factors. and pass-through to normal interest rate effects. For this reason, the term “credit channel” is often confusing; The credit channel is a strengthening mechanism, not an independent transmission channel like the interest rate channel.

In addition, unlike the traditional interest rate channel, the impact of monetary policy on the real economy through the balance sheet and bank lending channels, has important distributive results. Banks with different deposit dependence and businesses with different financial situation and dependence

depending on bank credit are not necessarily equally affected by monetary policy shocks.

2.1.2. Factors determining bank lending channel

Bernanke and Blinder (1988) argue that when central banks tighten monetary policy by selling securities on the open market, reserves and deposits of the banking system will decrease. Because it is not easy to issue debt instruments and shares to offset the decline in deposits, the ability of banks to provide credit decreases. Bernanke and Gertler (1995) argue that in economies with underdeveloped stock markets, banks remain the dominant source of credit, specializing in overcoming information problems and other barriers to entry. the credit market, so when the credit granting capacity of banks decreases, investment and consumption of entities in the economy will also decrease.

From theoretical concepts, Benkovskis (2008) believes that the bank lending channel can only operate effectively in the monetary policy transmission mechanism when the following two conditions are met: first, policy currency has the potential to affect the supply of bank credit (or deposits and bonds are not fully interchangeable for banks), and second, borrowers must depend on bank credit (or bank credit and bonds are not fully interchangeable for borrowers). Therefore, the importance of the bank lending channel is determined by two factors: first, the degree of impact of monetary policy on the supply of bank credit, and second, the degree of dependence of the borrowers. borrow into bank credit.

(i) The degree of impact of monetary policy on the supply of bank credit:

- Financial strength of the banking industry:

Cecchetti (1999) argues that the impact of monetary policy on the supply of bank credit depends on the financial strength of the banking industry. In general, the stronger a country's banking sector, the greater the impact of policy actions

is predicted to be weaker; The balance sheets of large, healthy banks are not policy sensitive because the decline in their risk reserves can easily be offset by other alternative sources of financing, so they usually do not care about the requirements for hedging.

There are various indicators to measure the financial strength of the banking industry in the research literature. Bank size, market concentration, capitalization, and liquidity are frequently mentioned factors (Cecchetti, 1999; Ehrmann et al., 2003). A bank's relatively small size, weak banking market concentration, liquidity and low capitalization would suggest a stronger lending channel, as such banks are more exposed to uncertainty. market conditions and will face more difficulties in attracting non-deposit financing. In addition, the financial strength of the banking industry is described by provisions for credit losses, operating costs, and return on assets, as well as the number of banks that have failed in the past (Benkovskis, 2008). ).

- Ownership structure of the banking industry:

Another important factor is the ownership structure of the banking industry in each country. State influence on the banking industry can be direct through bank ownership, state control, or through public guarantees by providing additional funding sources within the capacity of the bank. permission and reduce information asymmetry. Therefore, a country with relatively deep state involvement in the banking sector will undermine the importance of the bank lending channel in that country (Benkovskis, 2008).

Foreign participation in the domestic banking industry also weakens the bank lending channel, as foreign bank branches may experience less “financial constraints” due to the resources provided by the parent bank. additional funding as needed (De Haas and Naaborg, 2005; Ehrmann et al., 2003).

- Legal framework:

Kashyap and Stein (1993) argue that the impact of monetary policy on the supply of bank credit depends on the regulatory framework, since capital standards are based on

Risk exposure can limit a bank's ability to extend credit based on its equity and lending limits. On the other hand, a bank's lending behavior can also be affected by deposit insurance requirements, as a high level of deposit insurance reduces the risk exposure for customers. The lower the level of risk, the lower the cost of deposit mobilization for banks, thereby increasing the bank's dependence on these types of liabilities (liabilities).

- Bank credit term and bank interest rate:

Benkovskis (2008) states that the speed of money transmission depends on the bank credit term and the type of bank interest rate. The larger the market share of short-term credit with floating interest rates, the faster the response of credit supply to changes in monetary policy because these loans are quite sensitive to changes in interest rates. policy rate.

(ii) The borrower's dependence on bank credit:

Borrower 's financial situation:

The borrower's dependence on bank credit, often explained by the fact that banks play a special role in the financial system because they are well suited to deal with the problem of irregular information. asymmetry in credit markets (Mishkin, 1996). Borrowers with weak financial status, such as small and medium-sized enterprises and households, are most likely to suffer from asymmetric information problems, so their level of banking dependence is quite low. high. The higher the share of such borrowers in the credit market, the higher the bank dependency ratio and the more efficient the bank lending channel (Benkovskis, 2008).

- Availability of non-bank financial sources:

Bank dependence is also affected by the availability of non-bank financial resources. A country with an underdeveloped stock market, a relatively low capital market capitalization relative to bank assets and credit indicates higher bank dependence and a stronger bank lending channel in transmission. currency (Benkovskis, 2008).

2.2. Empirical evidence on bank lending channel

The topic of bank lending channels has received special attention from researchers over the past 20 years. The first empirical studies focused on macroeconomic evidence, since perhaps the simplest implication of the lending channel is that bank credit should be strongly correlated with indicators that are representative of the government. currency book. For example, Bernanke and Blinder (1988) found that an increase in the federal funds rate in the US prompted banks to lower credit growth. However, although the correlation between monetary policy indicators, bank credit, and economic activity is consistent with the view of the bank lending channel, such evidence still cannot provide a clear support for the lending channel. Kashyap and Stein (1993; 1995) argue that there is another way to explain these results, That is, the tightening monetary policy takes effect through the traditional interest rate channel, limiting economic activities and reducing credit demand. Therefore, there can be a convincing correlation between economic activity and bank credit even in the absence of a lending channel.

To overcome this identification problem, the use of disaggregated data on bank balance sheets has been proposed. If the view of the lending channel is correct, then the credit portfolios of banks with different characteristics will respond differently to a tightening in monetary policy (Kashyap and Stein, 1995, 2000). According to the lending channel theory, information imperfections in the financial markets have produced the effects of monetary policy on the supply of bank credit, as well as the resulting responses of credit supply differing over time. banks. The underlying assumption is that the more difficult it is for banks to offset the negative effects of tight monetary policy, the greater the extent to which it suffers from asymmetric information problems with its suppliers. capital for it is higher. The result is,

Two methods are commonly used in empirical studies to test the existence of a lending channel using disaggregated data of banks. One is to segment banks by bank characteristics (eg, Kashyap and Stein, 1995, 2000; Kishan and Opiela, 2000, 2006; Altunbas et al., 2002). However, using disaggregated data on bank balance sheets requires a large number of banks in the sample. This is no problem for the US and other countries with strong financial markets, but for countries with a small number of banks this approach is not feasible. Therefore, Ehrmann et al. (2003) proposed an alternative method that is to use the tabular model, allows the response of bank credit to changes in monetary policy to become dependent on bank-specific characteristics. This method is widely used in countries in Europe and Asia (eg, Benkovskis, 2008; Matousek and Sarantis, 2009; Hosono, 2006; Gunji and Yuan, 2010).

Many researchers have suggested several bank-specific characteristics to determine how sensitive different banks are to changes in monetary policy. Among them, the most commonly used characteristics in empirical studies on bank lending channels are size, liquidity, and bank capitalization (Kashyap and Stein, 1995, 2000; Benkovskis, 2008; Matousek and Sarantis, 2009).

- Bank size: Kashyap and Stein (1995, 2000) and Kishan and Opiela (2000) argue that small banks are more susceptible to information asymmetry problems than large banks. This will make small banks more sensitive to monetary policy shocks, since unlike large banks can easily increase non-deposit capital in response to money shocks. bad.

- Bank liquidity: Evidence provided by Kashyap and Stein (2000) and Ehrmann et al. (2003) shows that illiquid banks are more sensitive to changes in monetary policy , because the

Highly liquid banks can protect their credit portfolio by reducing liquid assets, while illiquid banks cannot do so.

Bank capitalization: Peek and Rosengren (1995) and Kishan and Opiela (2000, 2006) argue that banks with weak capital will reduce their credit supply more than well-capitalized banks capital after a monetary tightening, due to their limited access to non-deposit capital.

2.3. Previous studies on bank lending channel

2.3.1. Studies in the world

Influential research papers on the existence of bank lending channels in monetary policy transmission, using disaggregated data of banks, have been conducted in the US. Kashyap and Stein (1995) find that the growth of bank credit in the segment of small commercial banks is more responsive to monetary policy. Another study by Kashyap and Stein (2000) disaggregated banks not only by asset size, but also by liquidity. This study has shown that the smaller the banks, the more illiquid they are, the more responsive they are to monetary policy. Based on these results, the authors support the idea that there is a bank lending channel in monetary policy transmission in the US. Kishan and Opiela (2000) supported previous studies by disaggregating banks in terms of both size and capital strength. Kishan and Opiela (2006) study the asymmetric effects of monetary policy on the lending behavior of low- and high-capital banks, and the results are consistent with the predictions. on lending channels but only in the post-Basel era. Overall, banking studies in the US provide supporting evidence for the bank lending channel in monetary policy transmission, although this has recently been questioned by Ashcraft (2006). Using banking data, Ashcraft also found a differential response of credit supply to changes in Kishan and Opiela (2006) study the asymmetric effects of monetary policy on the lending behavior of low- and high-capital banks, and the results are consistent with the predictions. on lending channels but only in the post-Basel era. Overall, banking studies in the US provide supporting evidence for the bank lending channel in monetary policy transmission, although this has recently been questioned by Ashcraft (2006). Using banking data, Ashcraft also found a differential response of credit supply to changes in Kishan and Opiela (2006) study the asymmetric effects of monetary policy on the lending behavior of low- and high-capital banks, and the results are consistent with the predictions. on lending channels but only in the post-Basel era. Overall, banking studies in the US provide supporting evidence for the bank lending channel in monetary policy transmission, although this has recently been questioned by Ashcraft (2006). Using banking data, Ashcraft also found a differential response of credit supply to changes in although this has recently been questioned by Ashcraft (2006). Using banking data, Ashcraft also found a differential response of credit supply to changes in although this has recently been questioned by Ashcraft (2006). Using banking data, Ashcraft also found a differential response of credit supply to changes in

federal funds rate through banks. However, when we aggregate banking data at the state level, the credit market share of member banks (branches and subsidiaries) tends to mitigate the negative reaction of credit growth. state to changes in monetary policy, while the overall elasticity of output for bank lending is negative and insignificant.

There are also many empirical studies on bank lending channels in European countries. Matousek and Sarantis (2009) conducted an empirical study on the bank lending channel in money transmission in eight CEE countries that have joined the European Union: Czech Republic, Estonia, Hungary, Latvia, Lithuania, and Poland. Lan, Slovak Republic and Slovenia. Using panel data of a large number of banks for the period 1994-2003, and dynamic panel estimation techniques, the author found evidence of bank lending channels in all countries, although its strength varies from country to country. Bank size and liquidity seem to play the most important role in discriminating bank responses to changes in monetary policy. The author also studies the macroeconomic effects of the bank lending channel and finds evidence of a link between the overall credit supply and real economic activity in CEE countries. In general, however, the empirical evidence in Europe is less convincing. Favero et al. (1999) studied the existence of bank lending channels in Germany, France, Italy, and Spain during the monetary tightening period in 1992. They did not find any evidence. on bank lending channels in these countries. The study presented by De Bondt (1999) analyzed six European countries. These analyzes show that a bank lending channel exists in money transmission in Germany, Belgium, and the Netherlands when short-term interest rates are used as a proxy for monetary policy. No supporting evidence was found in France, Italy, and the UK. In the second part of the study, short-term interest rates are replaced by the Monetary Condition Index (MCI, which is an index used to measure the degree of easing or tightening of monetary policy), the bank lending channel. exists in monetary transmission in France and Italy. Ehrmann et al. (2003) research on the bank lending channel

rows using micro and aggregate data for Germany, France, Italy, and Spain. The authors find that illiquid banks are more responsive to changes in monetary policy than highly liquid banks, but overall bank size and capitalization are not important. important. Altunbas et al. (2002) also examined the evidence of bank lending channels in European countries. They found that across the Economic and Monetary Union (EMU) system, low-cap banks tend to be more responsive to changes in monetary policy, regardless of its size.

The existence of a bank lending channel in money transmission in the euro area has also been confirmed by numerous studies within individual countries. Benkovskis (2008) looked for evidence of the bank lending channel in Latvia by estimating a bank credit function that not only takes into account the monetary policy index and macroeconomic variables, but also calculates to bank-specific characteristics and differences in bank lending responses to monetary policy moves. The results show that a domestic currency shock has a distributive effect and that the affected banks are small, locally owned, and have low liquidity or capitalization. However, the bank lending channel is limited to credit supply in local currency only. this greatly reduces the importance of this channel in currency transmission in Latvia. Pruteanu (2004) discovered the existence of a bank lending channel in the money transmission mechanism in the period 1996 - 1998 in the Czech Republic, with capitalization dominates the impact of monetary policy on credit. bank. In addition, the emergence of liquidity also makes a difference in the response of bank credit to changes in monetary policy, but only for banks with a major participation of the government. country. Empirical analyzes conducted by Juks (2004) provide evidence in favor of the bank lending channel in Estonia, and the liquidity of Estonian banks seems to be an important determinant of bank lending. credit supply.

stronger than banks with more liquidity, although this evidence seems to be supported mainly by the effect of a credit portfolio composition. Loupias et al. (2003) report on some asymmetries between highly liquid and illiquid banks in France. Kakes and Sturm (2002) find that lending by small banks in Germany declines more than that of large banks after a monetary tightening. Gambacorta (2005) also finds similar evidence for banks in Italy regarding capitalization and liquidity.

Recently, the existence of bank lending channel has also been examined in Asian countries. Gunji and Yuan (2010) used bank-level data to investigate whether the impact of monetary policy on bank lending depends on the characteristics of Chinese banks over the period 1985 – 2007 or not. The authors found that the impact of monetary policy on bank lending was weaker for large banks and for banks with low liquidity, while bank capitalization was unlikely to have any plays an important role in discriminating the response of banks to changes in monetary policy. Hosono (2006) examines the existence of a bank lending channel in the monetary transmission mechanism in Japan using bank balance sheet data for the period 1975-1999.

2.3.2. Studies in Vietnam

In Vietnam, the topic of monetary policy transmission has attracted the attention of many researchers in recent years. However, these studies mainly provide an overview of the monetary policy transmission pathways in Vietnam in general, but do not focus on in-depth research on the bank lending channel in the main transmission. currency book. Le Viet Hung and Wade Pfau (2008) laid the foundation for quantitative studies on policy transmission

currency in Vietnam. The study uses a reduced VAR model to study the transmission channels in Vietnam in the period 1996 - 2005. The results of the study show that monetary policy has an impact on inflation and output. However, the statistical significance of the transmission channels is very weak, only the credit channel and the exchange rate channel are more statistically significant.

Tran Ngoc Tho (2013) studied the monetary policy transmission mechanism in Vietnam in the period 1998 - 2012 using the SVAR model. Interest rate, credit, and exchange rate pass-throughs were analyzed to determine the magnitude and duration of monetary policy transmission to the target variables. The results of the study show that inflation in Vietnam is more sensitive to the exchange rate channel than the other channels (especially the period after Vietnam's accession to the WTO). Nguyen Khac Quoc Bao et al (2013) use error correction vector model (VECM) method with variables being oil price, US base interest rate representing exogenous shocks, industrial output variable. , inflation, money supply, interest rates, nominal exchange rates and stock prices that represent the domestic economy, to examine the monetary policy transmission mechanism in Vietnam in the period 1999 - 2012. The results show that the degree of monetary policy transmission through the interest rate channel does not have a great influence on public output. industry, inflation, and the stock market when compared to other channels. The authors suggest that the above phenomenon may be due to the ineffective transmission from the policy interest rate to the market interest rate, leading to the ineffective use of the policy interest rate as in the traditional theoretical framework. system proposed. The results also show that the degree of monetary policy transmission through the credit channel (through money supply analysis) has a greater influence on industrial output, inflation, and the stock market than in other countries. remaining channel. From that,

important.

To find more evidence on the existence of credit channels in monetary policy transmission in Vietnam, Chu Khanh Lan (2013) conducted a study.

The experiment focuses on the bank credit channel. The author uses quarterly macroeconomic data for the period 2000 - 2011 and the VAR approach to assess the role of the credit channel in the transmission of monetary policy to the economy through measuring the relationship between output, inflation, interest rates, money supply and credit of the Vietnamese economy. The results of the study show that the amplification of the credit channel in the transmission of monetary policy is quite fast and strong in the Vietnamese economy. However, the study has not shown the asymmetric effects of monetary policy on credit growth across banks, therefore, the role of banks in policy transmission has not been clarified. currency through credit channels.

To the best of our knowledge, there is currently only one research paper addressing the topic of bank lending channels in Vietnam using bank microdata conducted by Nguyen Phuc Canh and Mr. et al (2013), in which the authors used the differential GMM method and panel data of Vietnamese commercial banks for the period 2003 - 2012 to analyze the role of banks in key transmission. monetary policy in Vietnam. Research results indicate that the capital, liquidity, and risk characteristics of banks will affect the response of bank credit to a currency shock. However, some limitations of this study should be acknowledged. The use of disproportionate panel data can lead to biased estimates due to sampling (Verbeek and Nijman, 1992), therefore, the authors should conduct a selection bias test to ensure the reliability of the estimated results. In addition, the study only stops at examining the impact of the currency shock on credit growth through banks, but has not yet measured the impact of credit growth on the macro variables of the banking system. economy to test whether monetary policy is really working. We will try to address these issues in our research paper.

CHAPTER 3

RESEARCH METHODS

3.1. Research models

Bernanke and Blinder (1992) argue that macro time series data are not useful in determining the bank lending channel - a sub-channel of the credit channel. Aggregated data do not allow us to distinguish between supply and demand factors that affect banks' lending. In contrast, disaggregated data for banks can effectively capture the distributional effects of monetary policy across the bank lending channel.

Two methods are commonly used in empirical studies to test the existence of a lending channel using disaggregated data of banks. One is to segment banks by size, capitalization, and liquidity (eg, Kashyap and Stein, 1995, 2000; Kishan and Opiela, 2000, 2006; Altunbas et al., 2002). This approach requires a large number of banks, which is no problem for the US. But the number of banks in Vietnam is relatively small, so this approach is not feasible. An alternative approach is to use a panel model, which allows the response of bank credit to changes in monetary policy to become dependent on banking characteristics (Ehrmann et al., 2003). ; Matousek and Sarantis, 2009). This approach helps to avoid the problem related to the number of banks mentioned above,

Matousek and Sarantis (2009) developed a credit market model drawn from Bernanke and Blinder (1988). Evidence for the bank lending channel is obtained by estimating a bank credit function that not only takes into account the monetary policy index and macroeconomic variables, but also takes into account specific characteristics. banks, which represent the response of bank credit to monetary policy directly (through the traditional interest rate channel, also known as the monetary channel) and to banking characteristics (the channel for get a loan). Similar to Matousek and Sarantis