《中国金融市场 Chinese Financial Markets》补充阅读文献：monetary policy and bank lending in china（2012-he dong）.pdf

Monetary Policy and Bank Lending in China evidence from loan level data 1. Introduction The transmission mechanism of monetary policy in China is little understood and largely black box to both researchers and policy makers. To a significant extent this is borrowers are influenced by not only regulated interest rates and quantity-based Y due to the So-called"dual-track "interest rates system, under which banks and the instruments, but also by signals contained in the variations of the market-determined interest rates(He and Wang, 2012). Such market-determined interest rates serve as shadow prices of loans, which would affect lending and borrowing decisions on the margin. Secondly, most of the dominant Chinese commercial banks are majority state- owned and may be required by policy makers to direct their loans to certain firms, which further complicate banks'lending behavior(Berger et al., 2010); Thirdly, very little micro-level data has been available publicly to allow researchers to examine bank lendi in China This paper attempts to shed light into the black box in two steps: First, a simple theoretical model is constructed to show how bank lending is affected by monetary policy under the dual-track "interest rates system and market frictions. Second, a proprietary hand-collected loan-level dataset enables us to examine bank lending behavior systematically. The panel dataset comprises of more than 11, 000 loan-level observations from 672 listed firms in the Shenzhen Stock Exchange(szse) from 2002 to 2011.In d dition we have detailed balance sheet data of these listed firms which allow us to control for firms' characteristics and identify how monetary policy and market interest rates affect bank lending The existing literature on links between monetary policy and bank lending is typically based on monetary transmission in advanced economies, in which policy changes from central banks can be transmitted to the real economy through interest rate channel and bank lending channel (bernanke and blinder, 1992; Bernanke and gertler, 1995; Mishkin, 1996; Kashyap and Stein, 2000; among others). However, since Chinese commercial banks face more regulations and market frictions, many basic questions remain to be answered. This paper aims to address the following questions: Do monetary policy instruments effectively influence bank lending in China? Do they affect the price and quantities of loans differently? Is bank lending affected by signals from non-regulated capital markets, and how important is the impact? Does the impact of monetary policy on 2 The reason we choose listed firms in Shenzhen Stock Exchange is to have better coverage of small and medium size firms: the representation of smaller firms in better on the Shenzhen Stock Exchange than on however,not all firms report their loan information correctly and completely(especially loan rafe ge: the Shanghai Stock Exchange. There are about 1,500 listed firms traded in Shenzhen Stock Excha

2 Monetary Policy and Bank Lending in China —Evidence from loan level data 1. Introduction The transmission mechanism of monetary policy in China is little understood and largely remains a black box to both researchers and policy makers. To a significant extent this is due to the so-called “dual-track” interest rates system, under which banks and their borrowers are influenced by not only regulated interest rates and quantity-based instruments, but also by signals contained in the variations of the market-determined interest rates (He and Wang, 2012). Such market-determined interest rates serve as shadow prices of loans, which would affect lending and borrowing decisions on the margin. Secondly, most of the dominant Chinese commercial banks are majority stateowned and may be required by policy makers to direct their loans to certain firms, which further complicate banks’ lending behavior (Berger et al., 2010); Thirdly, very little micro-level data has been available publicly to allow researchers to examine bank lending in China. This paper attempts to shed light into the black box in two steps: First, a simple theoretical model is constructed to show how bank lending is affected by monetary policy under the “dual-track” interest rates system and market frictions. Second, a proprietary hand-collected loan-level dataset enables us to examine bank lending behavior systematically. The panel dataset comprises of more than 11,000 loan-level observations from 672 listed firms in the Shenzhen Stock Exchange (SZSE) from 2002 to 2011.2 In addition, we have detailed balance sheet data of these listed firms, which allow us to control for firms’ characteristics and identify how monetary policy and market interest rates affect bank lending. The existing literature on links between monetary policy and bank lending is typically based on monetary transmission in advanced economies, in which policy changes from central banks can be transmitted to the real economy through interest rate channel and bank lending channel (Bernanke and Blinder, 1992; Bernanke and Gertler, 1995; Mishkin, 1996; Kashyap and Stein, 2000; among others). However, since Chinese commercial banks face more regulations and market frictions, many basic questions remain to be answered. This paper aims to address the following questions: Do monetary policy instruments effectively influence bank lending in China? Do they affect the price and quantities of loans differently? Is bank lending affected by signals from non-regulated capital markets, and how important is the impact? Does the impact of monetary policy on 2 The reason we choose listed firms in Shenzhen Stock Exchange is to have better coverage of small and medium size firms: the representation of smaller firms in better on the Shenzhen Stock Exchange than on the Shanghai Stock Exchange. There are about 1,500 listed firms traded in Shenzhen Stock Exchange; however, not all firms report their loan information correctly and completely (especially loan rate information)

bank lending vary with business cycles and overall stance of monetary policy, and do they differ across firms of different size? Previous studies on monetary policy in China mainly focus on the links between monetary policy and macroeconomic performance, and assume that the monetary transmission mechanism in China is the same as that in advanced economies(Qin et al (2005), Geiger(2006), Fan and Zhang(2007), Laurens and maino(2007); among others) A few recent studies discuss the transmission mechanism under regulated interest rates and find that monetary policy instruments are able to influence interest rates in money banking system)to the other(non-regulated money and bond markets)(Porter and Xy e and bond markets, suggesting policy signals can be transmitted from one track (regul 2009: Chen et al., 2011; He and wang, 2012) However, studies on the links between monetary policy and bank lending in China are rare. Bank lending, interest rate and asset price channels are found to exist in China(Sun et al., 2010), but the effectiveness of the bank lending channel varies across provinces and banks(Ho, 2012). More specifically, the impact of monetary policy on lending is weaker for larger banks and banks with lower levels of liquidity( Gunji and Yuan, 2010) However, none of these studies examines bank lending behavior using loan level data In this paper, we find that bank lending is not only affected by policy changes by the central bank, but also reacts to price changes in the non-regulated money and bond markets. Under the dual-track interest rates system, banks react to signals from both tracks when making loans The empirical analysis of the paper focuses on how the price and quantities of bank loan are affected by both monetary policy instruments and market-determined interest rates after controlling for loan features, bank types, macroeconomic variables and firm characteristics. We find that the People's Bank of China(PBc) can effectively influence loan rate through policy instruments, with the benchmark deposit rate being the most powerful instrument, followed by the Reserve Requirement Ratio(RRR). However, these two instruments have little impact on loan size. In contrast, both the price and quantity of loans are affected by the market-determined interest rates More interestingly, the impact from policy instruments and market interest rates would vary with firm size and the stance of monetary policy. While the effects on loan rates do not differ systematically across firms of different size, the effects on loan size are asymmetrical:, changes in benchmark deposit rate and market interest rates affect I firms more than smaller firms, but changes in RRR affect smaller firms more than firms The impact on loan rates from changes in the benchmark deposit rate and the rrr would also vary with the prevailing monetary policy stance. Changes in the benchmark deposit rate have larger effects on loan rates when monetary policy is being loosened than when Hereafter, the market interest rate means the representative interest rates(such as the 7-day Repo rate)in non-regulated money and bond markets

3 bank lending vary with business cycles and overall stance of monetary policy, and do they differ across firms of different size? Previous studies on monetary policy in China mainly focus on the links between monetary policy and macroeconomic performance, and assume that the monetary transmission mechanism in China is the same as that in advanced economies (Qin et al (2005), Geiger (2006), Fan and Zhang (2007), Laurens and Maino (2007); among others). A few recent studies discuss the transmission mechanism under regulated interest rates and find that monetary policy instruments are able to influence interest rates in money and bond markets, suggesting policy signals can be transmitted from one track (regulated banking system) to the other (non-regulated money and bond markets) (Porter and Xu, 2009; Chen et al., 2011; He and Wang, 2012). However, studies on the links between monetary policy and bank lending in China are rare. Bank lending, interest rate and asset price channels are found to exist in China (Sun et al., 2010), but the effectiveness of the bank lending channel varies across provinces and banks (Ho, 2012). More specifically, the impact of monetary policy on lending is weaker for larger banks and banks with lower levels of liquidity (Gunji and Yuan, 2010). However, none of these studies examines bank lending behavior using loan level data. In this paper, we find that bank lending is not only affected by policy changes by the central bank, but also reacts to price changes in the non-regulated money and bond markets. Under the dual-track interest rates system, banks react to signals from both tracks when making loans. The empirical analysis of the paper focuses on how the price and quantities of bank loans are affected by both monetary policy instruments and market-determined interest rates after controlling for loan features, bank types, macroeconomic variables and firm characteristics. We find that the People’s Bank of China (PBC) can effectively influence loan rate through policy instruments, with the benchmark deposit rate being the most powerful instrument, followed by the Reserve Requirement Ratio (RRR). However, these two instruments have little impact on loan size. In contrast, both the price and quantity of loans are affected by the market-determined interest rates.3 More interestingly, the impact from policy instruments and market interest rates would vary with firm size and the stance of monetary policy. While the effects on loan rates do not differ systematically across firms of different size, the effects on loan size are asymmetrical: ,changes in benchmark deposit rate and market interest rates affect larger firms more than smaller firms, but changes in RRR affect smaller firms more than larger firms. The impact on loan rates from changes in the benchmark deposit rate and the RRR would also vary with the prevailing monetary policy stance. Changes in the benchmark deposit rate have larger effects on loan rates when monetary policy is being loosened than when 3 Hereafter, the market interest rate means the representative interest rates (such as the 7-day Repo rate) in non-regulated money and bond markets

it is being tightened, but chanages in the rrr have larger effects on loan rates when monetary policy stance in being tightened than when it is loosened. In addition, changes the market interest rates have larger effect on loan size when being loosened than when it is being tightened The rest of the paper is organized as follows. The next section briefly reviews the institutional background of the Chinese monetary policy framework and the banking sector. Section 3 develops a simple theoretical model and discusses its predictions Section 4 describes the specifications of empirical models and the estimation strategy Section 5 describes the data and discusses sample selection. Section 6 reports the estimation results and discusses two caveats. Section 7 concludes the paper 2. Institutional Background 2.1 The monetary policy framework in China According to the Law on the People's Bank of China, "the aim of monetary policies shall be to maintain the stability of the currency and thereby promote economic growth. Thus the pbc has a dual mandate. similar to that of the us Federal reserve. Even though it is not explicitly stated in the law, there is also an understanding that the PbC has the mandate to maintain the stability of the Chinese financial system, reflecting its role as the lender of last resort. The policy implementation framework has evolved since the mid-1990s from relying on quantity-based instruments into a mixture of both quantity and price-based instruments. (He and Pauwels, 2008) Key to a good understanding of China's monetary policy framework is the"dual-track interest-rate system: on the one hand, bank deposit and lending rates are regulated by the central bank (i. e, the imposition of a deposit-rate ceiling and a lending- rate floor ) on the other hand, interest rates in money and capital markets are market-determined. At the same time, the ceiling or the floor may not necessarily be binding in practice. The deposit-rate ceiling is generally considered binding while actual lending data since 2004 suggests the lending- rate floor is generally not binding(He and wang, 2012) Price distortions (rates ceiling and floor)in the banking system mean that the pbc m also rely on quantity-based instruments to achieve its targets. For example, a lower Qust deposit rate ceiling(compared to its equilibrium level) causes the loan supply curve of commercial banks to shift to the right(Graph 1, S1>S2), where Si is the loan supply curve without the deposit rate ceiling. The shifted loan supply curve(S2)means that banks are willing to lend to firms at lower loan rates because the funding cost of banks (deposit rate) is lower than it should be(P2<Pl). Meanwhile, firms' loan demand is higher than its equilibrium level due to cheaper rates( Q2>Q1). This means that both commercial banks and firms benefit from lower loan rates but at the expense of depositors in the economy who receive lower rates on their deposits However, the new equilibrium under the deposit rate ceiling(P2, Q2)means that there will be more credit(Q2>Q1)in the economy compared to its original equilibrium(Pl

4 it is being tightened, but chanages in the RRR have larger effects on loan rates when monetary policy stance in being tightened than when it is loosened. In addition, changes in the market interest rates have larger effect on loan size when monetary policy stance is being loosened than when it is being tightened. The rest of the paper is organized as follows. The next section briefly reviews the institutional background of the Chinese monetary policy framework and the banking sector. Section 3 develops a simple theoretical model and discusses its predictions. Section 4 describes the specifications of empirical models and the estimation strategy. Section 5 describes the data and discusses sample selection. Section 6 reports the estimation results and discusses two caveats. Section 7 concludes the paper. 2. Institutional Background 2.1 The monetary policy framework in China According to the Law on the People’s Bank of China, “the aim of monetary policies shall be to maintain the stability of the currency and thereby promote economic growth.” Thus, the PBC has a dual mandate, similar to that of the US Federal Reserve. Even though it is not explicitly stated in the law, there is also an understanding that the PBC has the mandate to maintain the stability of the Chinese financial system, reflecting its role as the lender of last resort. The policy implementation framework has evolved since the mid-1990s from relying on quantity-based instruments into a mixture of both quantityand price-based instruments. (He and Pauwels, 2008). Key to a good understanding of China’s monetary policy framework is the “dual-track” interest-rate system: on the one hand, bank deposit and lending rates are regulated by the central bank (i.e., the imposition of a deposit-rate ceiling and a lending-rate floor); on the other hand, interest rates in money and capital markets are market-determined. At the same time, the ceiling or the floor may not necessarily be binding in practice. The deposit-rate ceiling is generally considered binding while actual lending data since 2004 suggests the lending-rate floor is generally not binding (He and Wang, 2012). Price distortions (rates ceiling and floor) in the banking system mean that the PBC must also rely on quantity-based instruments to achieve its targets. For example, a lower deposit rate ceiling (compared to its equilibrium level) causes the loan supply curve of commercial banks to shift to the right (Graph 1, S1S2), where S1 is the loan supply curve without the deposit rate ceiling. The shifted loan supply curve (S2) means that banks are willing to lend to firms at lower loan rates because the funding cost of banks (deposit rate) is lower than it should be (P2<P1). Meanwhile, firms’ loan demand is higher than its equilibrium level due to cheaper rates (Q2>Q1). This means that both commercial banks and firms benefit from lower loan rates but at the expense of depositors in the economy who receive lower rates on their deposits. However, the new equilibrium under the deposit rate ceiling (P2, Q2) means that there will be more credit (Q2>Q1) in the economy compared to its original equilibrium (P1

Q1), which might conflict with the PBC's inflation target. In order to prevent a deviation from its inflation target, the PBC will have to constrain the credit supply in the economy At least two measures are introduced for this purpose: First, a lending rate floor is used to reduce loan demand from firms through higher loan rates(P3>P2). This floor, coupled with the deposit rate ceiling, also means that a decent profit margin is guaranteed for banks in the sector. Second, multiple quantity-based instruments such as a credit quota and the rrR are introduced to restrain credit supply in the banking system in order to keep inflation low in the economy From the simple analysis above, we can see that distortions caused by price regulations have to be corrected by quantity-based instruments, which is why the Pbc has to use both price-based and quantity-based instruments in its framework. On the other hand money and bond markets have also emerged since the 1990s, where interest rates have been largely determined by market forces. Commercial banks can choose to borrow and lend in money and bond markets whilst they are subject to regulated interest rates in the banking sector. Understanding the mechanisms of such a dual-track interest rate system is essential to an understanding of the links between monetary policy and bank lending Lower deposit SI loan supply S2 PI Graph 1: loan market in banking system 2.2 The Chinese banking sector The financial system in China has gone through a fundamental structural transformation since the late 1990s. Although bank credit is still the dominant form of financial intermediation, off-balance sheet activities(such as trust and entrusted loans)and market- based financial intermediation have grown very fast in recent years. Indeed, the share of bank credit in total social financing, a measure of the aggregate volume of financial intermediation through both the banking sector and capital markets, had fallen from more than two thirds in early 2000s to under 50% in the second half of 2011(PBC, 2012)

5 Q1), which might conflict with the PBC’s inflation target. In order to prevent a deviation from its inflation target, the PBC will have to constrain the credit supply in the economy. At least two measures are introduced for this purpose: First, a lending rate floor is used to reduce loan demand from firms through higher loan rates (P3>P2). This floor, coupled with the deposit rate ceiling, also means that a decent profit margin is guaranteed for banks in the sector. Second, multiple quantity-based instruments such as a credit quota and the RRR are introduced to restrain credit supply in the banking system in order to keep inflation low in the economy. From the simple analysis above, we can see that distortions caused by price regulations have to be corrected by quantity-based instruments, which is why the PBC has to use both price-based and quantity-based instruments in its framework. On the other hand, money and bond markets have also emerged since the 1990s, where interest rates have been largely determined by market forces. Commercial banks can choose to borrow and lend in money and bond markets whilst they are subject to regulated interest rates in the banking sector. Understanding the mechanisms of such a dual-track interest rate system is essential to an understanding of the links between monetary policy and bank lending. Graph 1: loan market in banking system 2.2 The Chinese banking sector The financial system in China has gone through a fundamental structural transformation since the late 1990s. Although bank credit is still the dominant form of financial intermediation, off-balance sheet activities (such as trust and entrusted loans) and marketbased financial intermediation have grown very fast in recent years. Indeed, the share of bank credit in total social financing, a measure of the aggregate volume of financial intermediation through both the banking sector and capital markets, had fallen from more than two thirds in early 2000s to under 50% in the second half of 2011 (PBC, 2012). P Q Q1 Q2 S2 S1 P1 P2 D Lower deposit ceiling leads loan supply curve to shift to right Lending rate P3 floor