商业银行净利差决定因素的实证探讨
Determinants of commercial bank net interest margin of empirical study
净利差(Net Interest Margins, NIM)不仅是商业银行利润的重要来源，而且是衡量商业银行效率的重要指标(Demirgiiu－Kunt, 1999)，越来越被学界关注。净利差反映金融中介在资源配置中的交易成本，高利差往往意味着效率损失(Saunders and Schumacher, 2000)。在我国利率市场化改革不断深化的背景下，研究净利差的决定因素以及各因素的相对重要性，可以揭示出在不确定的市场环境中影响商业银行存贷款定价的因素，不仅有助于提高商业银行在利率市场上的定价能力并改善经营管理水平，而且也有助于金融监管当局更加全面的了解本国银行业的发展水平，进而实施有效的利差管理，提高监管效率。本文主要从理论和实证两个方面对银行净利差决定因素研究的最新进展进行梳理。
Net Interest margin (Net Interest Margins, NIM) is not only a commercial Banks an important source of profits, and is one of the important indicators to measure commercial bank efficiency (Demirgiiu - Kunt, 1999), more and more academic attention. Net interest margin reflect financial intermediation in the allocation of resources and the transaction cost, high spreads often means efficiency loss (Saunders and Schumacher, 2000). Under the background of China's market-oriented interest rate reform continued to deepen, the determinants of the net interest margin and the relative importance of various factors, can be revealed in an uncertain market environment factors that influence the commercial bank deposit and lending pricing will not only help improve the ability of commercial Banks in interest rates on the market pricing and improve the management level, but also contribute to a more comprehensive understanding of the financial regulatory authorities, the development level of domestic banking and effective spreads management, improve the regulation efficiency. This article mainly from two aspects of theory and practice for bank net interest margin sorted out the latest progress in the study of decision factors.
一、银行净利差的测算
A measure, a bank's net interest margin
自20世纪70年代末开始的金融自由化浪潮以来，利率在金融市场中的资源配置作用日益重要，关于存贷款之间的净利差问题也自然引起了学术界的广泛讨论，净利差的定义目前学界并未取得一致看法。①Ho & Saunders(1981)把净利差等同为价差(Spread)或一种溢价(Mark－up)，即等于(利息收入/银行资产)-(利息支出/银行负债)，此后的研究者如Mcshane & Sharpe(1985)，Allen(1988)等也基本遵从了这一界定。但Angbazo(1997)指出，虽然净利差和价差都能反映出银行管理信贷风险和利率风险的能力，但并不完全一致，他认为与spread的计算不同，NIM应理解为(利息收入-利息支出)/生息性资产。Demirgiiu－Kunt & Huizinga(1999)则进一步对事前价差(Exante Spreads)和事后价差(Expost Spreads)做出了区分，认为考虑了违约风险的事后价差更适合于衡量银行效率。Brock & Suarez(2000)在研究中更是系统列出了三组狭义和广义上关于净利差的测算公式，经过对比分析后认为，(利息收入+佣金收入)/所有生息资产-(利息支出+佣金支出)/所有生息负债，②更能反映出金融中介的真实交易成本。
Since the late 1970 s started since a wave of financial liberalization, interest rate in the resource allocation function of financial market is becoming more and more important, the net interest margin between deposit and lending problems naturally caused academia widely discussed, the definition of net interest margin at present academic circles did not agree. 1) Ho & Saunders (1981), the net interest margin equal to Spread (Spread) or a premium (Mark - up), which is equal to (bank interest income/assets) - (interest payments/bank liabilities), and since then the researchers such as Mcshane & Sharpe (1985), Allen (1988) also basic conform to this definition. But Angbazo (1997) pointed out that, although the net interest margin and spreads can reflect the bank's ability to manage credit risk and interest rate risk, but is not entirely consistent, he thought with the spread of different calculations, NIM should be understood as interest income - interest payments)/interest-earning assets. Demirgiiu - Kunt & Huizinga (1999) further spread to advance (Exante Spreads) and after the event Spreads (Expost Spreads) made a distinction, that takes into account the risk of default of ex post price more suitable for measuring efficiency of Banks. Brock & Suarez (2000) in research system are listed in three groups on the narrow and broad formula about the calculation of net interest margin, through comparative analysis, concluded that + commission income (interest income)/all interest-bearing assets - (interest payments plus commission payments)/all interest-bearing debt, (2) can reflect the real financial intermediary transaction costs.
二、银行净利差决定的理论模型
Second, the bank's net interest margin decision theory model
银行净利差决定的理论模型归纳起来主要有两种：一种是Ho & Saunders(1981)提出的交易者模型(Dealer Model)，后经Allen(1988)等扩展并完善；另一种是基于Klein(1971)等基础上的银行公司微观模型(Micro－Model of the Banking Firm)。
Bank net interest margin decision theoretic Model sum up basically has two kinds: one kind is Ho & Saunders (1981) Model proposed by traders (Dealer Model), after approved by Allen (1988) extended and perfected; Another is based on Klein (1971), Banking Firm on the basis of microscopic Model (Micro - the Model of the Banking Firm).
(一)Ho-Saunders的交易者模型及其扩展
(a) Ho - Saunders traders model and its extension
-Saunders的交易者模型(Dealer Model)
1. Ho - Saunders traders Model (Dealer) Model
虽然Samuelson早在1945就指出了利率波动与银行资产组合之间关系的重要性，但直到1981年，Ho & Saunders系统提出交易者模型的理论框架，这才为此后关于银行净利差决定因素的研究奠定了理论基础。该模型假定银行作为一个风险厌恶的交易商，在一个简单的单期模型中，即使期初设定好存贷款利率，但由于贷款需要的资金和存款提供的资金在时间上存在着不匹配性，这使得银行不得不面临一定的利率波动风险。因此，银行必须在满足流动性的基础上，选择恰当的存贷款利率来减少由于存款供应与贷款需求不匹配而带来的损失，以实现效用最大化。其模型经推导得出如下形式：
Although Samuelson as early as 1945 points out the importance of the relationship between interest rates and bank asset portfolio, but it was not until 1981, Ho & Saunders, the system framework of traders model theory is put forward, this just about bank net interest margin for later study of the determinants of laid a solid theoretical basis. The model assumes that the bank as a risk aversion of the trader, in a simple single phase model, even if the initial deposit and lending rates set, but are still in need of loan, cash and deposits on time there is a mismatch of capital, which Banks have to face some risk of interest rate fluctuations. Banks must, therefore, to meet liquidity, on the basis of choosing the proper deposit and lending rates to reduce since the supply of deposits and loan demand does not match and losses, in order to achieve maximum utility. Its model by the derived form is as follows:
其中，s表示净利差，a和b是对银行为公众提供及时性存贷款服务而面临利率风险的补偿。第一部分α/β代表银行在完全风险规避时的利差，即纯利差。较大的α和较小的β，即α/β较大，意味着银行在市场中面临着相对无弹性的贷款需求和存款供给函数，也就是说银行在市场中处于垄断地位。银行在市场中垄断势力越强，净利差就越大，反之，当市场竞争程度越高时(低α/β)，净利差就越小。第二部分是对第一部分纯利差的风险调整，取决于三个因素：(1)R表示银行管理者的风险厌恶程度；(2)Q表示银行交易规模大小；(3)表示存贷款利率的瞬间波动。第二部分表明，在其他条件不变时，银行管理者的风险厌恶程度越高、银行的交易规模越大、利率波动越大，银行的净利差就越高。因此该模型表明，净利差是银行交易面临着不确定性的补偿，其大小取决于四个因素：市场结构、管理者厌恶风险程度、交易规模和利率波动。
Among them, s net interest margin, a and b is for Banks to offer deposit and loan services while facing the public interest rate risk compensation. The first part of alpha/beta, on behalf of the bank spreads when complete risk aversion, namely net profit difference. Larger smaller alpha and beta, alpha/beta, means that Banks face in the market is relatively inelastic demand for credit and deposit supply function, that is to say, the bank in a monopoly position in the market. Bank, the stronger the monopoly power in the market, the greater the net interest margin, on the contrary, when the higher level of competition in the market (low alpha/beta), the less the net interest margin. The second part is the difference in net profit for the first part of the risk of adjustment, depends on three factors: (1) R said bank managerial risk aversion; (2) Q said bank transaction size; (3) the said deposit and lending rates of instantaneous volatility. Part 2 shows that, when other conditions remain unchanged, the bank manager of the scale of the higher level of risk aversion, bank business, interest rates, the greater the volatility, the higher net interest margin of Banks. So the model shows that the net interest margin is bank transactions is facing uncertainty compensation, its size depends on four factors: market structure, management degree of risk aversion and transaction scale and interest rates.
-Saunders模型的扩展
2. Ho - Saunders, the extension of the model
Allen(1988)进一步放宽了Ho-Saunders模型中贷款同质性的假设，认为银行贷款产品或服务是多样化的，应该考虑存在贷款异质性(Heterogeneity)条件下对银行净利差的组合效果。通过模型推导，认为银行存在不同类型的贷款产品或服务时，多种贷款产品之间的竞争性，即不同贷款项目之间的需求交叉弹性将会导致银行利差的降低。该研究为此后银行信贷业务应当多元化和提升非利息收入比重等发展方向提供了理论模型上的支持。
Allen (1988) further relax the Ho - Saunders loan homogeneity assumption in the model, products or services that bank loans is diverse, and Heterogeneity of existing loans (Heterogeneity) should be considered under the condition of the combination of the bank's net interest margin. Through the model deduction, thinks that there are different types of bank loan products or services, a variety of loan products, the competitive between the cross elasticity of demand between different loan program will lead to a drop in bank spreads. The study of bank credit business should develop diversified and increase the proportion of non-interest income direction provides a theoretical support on model.
Angbazo(1997)沿着交易者模型的分析框架进一步考虑了信用风险和利率风险对银行最优净利差的影响。模型认为商业银行会倾向于要求更大的净利差以补偿高违约率和利率波动所带来的风险。最优利差模型表明，银行的净利差由市场结构、银行管理者的风险厌恶程度、交易规模和存贷行为的风险决定，而银行存贷行为的风险又由信用风险、货币市场利率风险以及信用风险和利率风险交叉影响决定。如果一家银行面临较高的信用风险和利率风险的话，会选择较高的贷款利率，从而获得较高的利差。
Angbazo (1997) along the traders model analysis framework to further consider the credit risk and interest rate risk on Banks' optimal net interest margin. Model that commercial Banks will tend to demand more net interest margin to compensate for the high default rate and the risk of interest rate fluctuations. Optimal spreads model shows that the net interest margin of Banks by market structure, bank managers' degree of risk aversion and transaction scale and the risk of loan behavior decision, risk and the risk of bank loan behavior by credit, money market interest rate risk and credit risk and interest rate risk cross influence decisions. If a bank faces a higher interest rate risk, credit risk and choose higher lending rates, thus higher spreads.
Maudos & Guevara(2004)又更进一步将商业银行运营成本纳入分析框架。在考虑了银行运营成本后，该模型认为即使商业银行市场是完全竞争的，而且不考虑信用和利率风险，商业银行也需要一定的净利差用以补偿其运营成本，这为银行净利差的存在做了另一层面的解释。
Maudos & Guevara (2004) and further commercial Banks operating costs should be brought into the analysis framework. After considering the bank operating costs, the model would think even commercial bank market is perfectly competitive, and no account of credit and interest rate risk, commercial Banks also need certain net interest margin in order to compensate for the operational cost, it has done for the existence of the bank's net interest margin another level of explanation.
为了解释商业银行现代业务对银行净利差的影响，Valverde & Fernández(2007)在Allen(1988)异质性产品模型的基础上进一步把银行业务区分为传统的存贷款业务和其他非传统业务(中间业务)，并在模型中引入动态性，认为商业银行净利差的影响因素除了市场结构、管理者的风险厌恶程度、交易规模、利率波动、银行不同产品的需求交叉弹性之外，中间业务活动会明显改变银行收入结构，即降低了净利差，增加了银行非利息收入。
To explain modern business of commercial bank's influence on the bank's net interest margin, Mr Valverde & ms Fernandez in Allen (2007) (1988) heterogeneous product model on the basis of further put the banking business is divided into the traditional deposit and lending business and other non-traditional business business (middle), and dynamic, introduced in the model, thought the influence factors of commercial bank's net interest margin in addition to market structure, managerial risk aversion and transaction scale, interest rate fluctuation cross elasticity of demand of different products, Banks, intermediary business activity can obviously change the bank income structure, which reduces the net interest margin, increased bank non-interest income.
Maudos & Solís(2009)在研究墨西哥的银行系统时建立了一个综合模型，在Ho－Saunders(1981)理论模型的基础上，将Maudos & Guevara(2004)模型中的运营成本和Valverde & Fernández(2007)中的非传统业务两个因素同时纳入了分析框架，以对银行净利差的影响进行分析。研究认为银行净利差除了受到信贷和利率风险影响外，更主要的是受到银行市场垄断势力、运营成本和非传统业务的影响，市场垄断势力越强、银行运营成本越高，则银行会要求更多的净利差，而非传统业务发展则与银行净利差呈反向关系。
Maudos & echoed to the Mexico (2009) in the banking system has set up a comprehensive model, in Ho - Saunders (1981), on the basis of theoretical model will be Maudos & Guevara (2004) model of the operating costs and Mr Valverde & ms Fernandez (2007) of non-traditional business two factors into the analysis framework at the same time, in order to analyze the influence of the bank's net interest margin. Research think bank net interest margin in addition to the affected by the credit and interest rate risk, more major is banking market monopoly power, the influence of the operating costs and non-traditional business, the stronger the market monopoly power, bank operation cost is higher, the bank will require more of the net interest margin, whereas non-traditional business development and inverse relationship to the bank's net interest margin.
(二)银行公司微观模型(Micro－model of the Banking Firm)
(2) bank company microscopic model (Micro - the model of the Banking Firm)
与前面介绍的Ho et al.基于Stoll(1978)的买卖价差(Bid－Ask Prices)模型不同，Zarruck(1989)，Zarruk & Madura(1992)，Wong(1997)等在Klein(1971)等提出，银行公司微观模型的基础上，根据厂商效用最大化原则，认为在一个静态的环境中银行作为中介机构，其主要作用是使存贷款的需求和供给同时达到市场均衡。
As the previous introduction of Ho et al. Based on the is (1978) bid-offer spreads (the Bid - Ask Prices) model is different, Zarruck (1989), Zarruk & Madura (1992), Wong (1997) in Klein (1971) put forward, such as bank company, on the basis of microscopic model, according to the manufacturer's utility maximization principle, thought in a static environment Banks as intermediaries, its main role is to make the deposit and loan demand and supply at the same time to achieve market equilibrium.
Zarruk(1989)认为，在风险厌恶的合理假设下，银行净利差与银行资本数量正相关，即随着核心资本增加而增加，但却随着银行存款的波动而减少。③Zarruk & Madura(1992)进一步认为，存款准备金和存款保险的要求会对银行净利差产生负面影响。由于银行资产质量与净利差成正向关系，贷款损失增加意味着银行资产质量下降，因而会对净利差有负面影响。
Zarruk (1989) argue that, in the reasonable assumption that risk aversion, the bank's net interest margin and the number of bank capital, namely capital increases as the core, but with the volatility of bank deposits. (3) Zarruk & Madura (1992) further argues that bank reserve against deposits and deposit insurance will have negative impact on bank's net interest margin. As Banks asset quality and has a positive relationship net interest margin, loan losses increase means that Banks asset quality drop, thus will have a negative influence on net interest margin.
与Zarruk(1989)和Zarruk & Madura(1992)只分析某一风险影响的单因素模型不同，Wong(1997)建立了一个在多重不确定性和风险厌恶条件下银行最优NIM决定理论模型，该模型不是对前者的简单延续，而是与Ho等人的交易者模型具有更多的一致性，可以认为是对两大模型兼容的一种尝试。该模型证明NIM与银行风险规避程度、市场垄断势力、运营成本、面临的利率风险和信用风险等因素呈正向关系，同业拆借利率的影响取决于该银行在拆借市场中的位置；当利率风险不明显时，对银行资本金的要求会减少净利差，否则影响将不能确定，而在同等条件下激励问题的引入则会降低银行净利差水平。
With Zarruk (1989) and Zarruk & Madura (1992) analysis of the influence of a particular risk only single factor model is different, Wong (1997) established a under the condition of multiple uncertainty and risk aversion bank optimal NIM decision theory model, this model is not a simple continuation of the former, but traders from Ho and others model is more consistent, can be thought of as for two models compatible a try. The model proves that NIM and bank, market monopoly power, operating costs by the degree of risk aversion, the interest rate risk and credit risk factors were positively to the relationship, the interbank rate depends on the influence of the Banks in the lending market position; When the interest rate risk is not obvious, will reduce net interest margin of bank capital requirements, otherwise the effect will not be able to determine, and the introduction of the incentive problem under the same conditions, would reduce the bank's net interest margin level.
三、银行净利差决定因素的实证研究
: an empirical study of three determinants, bank net interest margin
关于商业银行净利差决定因素的实证研究，主要是以交易者模型为理论基础展开的，我们将分地区进行讨论。
Determinants of commercial bank's net interest margin of empirical research, mainly traders model as the theoretical basis, we will discuss.
(一)对欧美发达国家和地区银行的研究
(a) the study of developed countries and regions in Europe and the bank
Ho & Saunders(1981)运用构建的交易者模型，并以美国商业银行1976－1979年13个季度数据为研究对象，实证研究表明由于银行在交易过程中面临着不确定性，净利差总是存在的，且与利率的波动性正相关；若把所观察的银行分成大银行(Large Money Center Banks)和小银行(Regionally Oriented Banks)两组，发现小银行的净利差高于大银行，原因在于大小银行的市场结构不同，大银行组中的竞争性更强，相对而言小银行在当地更具有垄断性。McShane & Sharpe(1985)根据澳大利亚的情况，对该模型进行了修正，实证研究认为银行的市场力量越大，银行的利差也就越大，银行的利差与绝对风险规避程度、市场利率变动的不确定性和平均交易量存在正相关性。
Ho & Saunders (1981) using the building trader model, and commercial Banks in the United States 13 quarterly data from 1976 to 1979 as the research object, the empirical research shows that as Banks face uncertainty in the process of trading, the net interest margin is always present, and positively related to the volatility of interest rates; If the observations into big bank, Money Center Banks) and smaller Banks (Regionally Oriented Banks) in the two groups, found the little bank's net interest margin is higher than the big Banks, because the size of bank market structure, the big Banks more competitive in the group, the relatively smaller Banks at a local monopoly. McShane & Sharpe (1985) according to the Australian situation, has carried on the correction to the model, the empirical research that bank market power, the greater the greater the bank interest rate differentials, bank interest rate differentials with the absolute risk aversion, the uncertainty of changes in market interest rates and the average trading volume is positive correlation.
Angbazo(1997)用美联储Call Report数据，对1989－1993年不同规模的286家银行进行研究，表明商业银行的净利差与违约风险和利率风险溢价有关。银行净利差与资本充足率、存款准备金的机会成本、资产管理水平、违约风险、隐含利息支付(Non-Interest Bearing Reserves)正相关，与流动性风险、贷款限制、利率风险负相关。同时Angbazo将美国商业银行划分为货币中心银行(Money-Center Bank)、跨区域银行(Super-Regional Bank)、区域性银行(Regional-Bank)和地方性银行(Local Bank)四类，④结果表明随着银行规模的不同，信用风险和利率风险的影响也不同。货币中心银行的净利差率与违约风险正相关，但与利率风险关系不明显，⑤这与货币中心银行更关注短期资产、用远期和表外工具对冲利率风险损失有关。而与此相反，跨区域银行的净利差对利率风险敏感但对违约风险不敏感，地方小银行的净利差则对违约风险和利率风险都比较敏感。
Angbazo (1997) with the fed's Call Report data, for different size of 286 Banks in 1989-1993, shows that the net interest margin of commercial Banks related to default risk and interest rate risk premium. Bank net Interest margin and capital adequacy ratio, the opportunity cost of Reserves, asset management, risk of default, implicit Interest payments (Non - Interest Bearing Reserves), and liquidity risk, credit limit, Interest rate risk. While Angbazo . commercial Banks are divided into Money Center Banks (Money Center Bank), across the region's Banks (Super - Regional Bank), Regional Banks (Regional Bank and Local Bank (Local Bank) four categories, (4) the results show that as Banks scale is different, the effect of credit risk and interest rate risk is also different. Money center bank net interest rate and default risk are related, but not obvious relationship with the interest rate risk, (5) with the money center Banks are more focused on the short-term assets, with forward and off-balance sheet vehicles to hedge interest rate risk loss. By contrast, cross regional bank's net interest margin is sensitive to interest rate risk but is not sensitive to the risk of default, the bank's net interest margin is small places on default risk and interest rate risk are sensitive.
利用7个OECD国家1988－1995年银行数据，Saunders & Schumacher(2000)分析认为，银行净利差与隐含利息支付、储备金的机会成本、资本充足率、市场垄断势力、利率波动正相关。其中，美国商业银行的地域和行为限制最严，净利差就越高，相对而言英国和法国银行间的竞争最激烈。利率波动会显著增大净利差。如果宏观政策能减少利率波动，那净利差就会降低。
Using seven OECD countries bank data from 1988 to 1995, Saunders & Schumacher (2000) analysis, the bank's net interest margin and implicit interest payments, the opportunity cost of reserves, capital adequacy and market monopoly power, interest rate volatility are related. Among them, the region and the behavior of commercial Banks limit the most severe, the higher net interest margins, relatively the most intense competition between the British and French Banks. Interest rate volatility will significantly increase the net interest margin. If macro policy can reduce the interest rate fluctuations, it will reduce net interest margin.
首次将运营成本纳入分析框架后，Maudos & Guevara(2004)分析了主要欧洲国家(德、法、英、意和西班牙)1993－2000年银行数据，研究表明市场垄断势力对银行净利差有显著影响。市场垄断势力下降和集中度降低，银行利差将下降；同时发现，相对于利率风险与信用风险，欧盟银行的净利差更多的受管理质量、产品成本、市场垄断势力影响；尤其是运营成本与净利差成正比，平均运营成本下降10％，净利差将会减少43％。因此在这段时间内，运营成本是影响净利差最为重要的因素之一。
Operating costs should be brought into the analysis framework for the first time, after Maudos & Guevara (2004) analysis of the major European countries (Germany, France, Britain, Italy and Spain) bank data from 1993 to 2000, the study suggests that the market monopoly power has a significant influence on the bank's net interest margin. Market monopoly power decline and concentration are low, Banks spreads will decline; Also found that relative to the interest rate risk and credit risk, the bank's net interest margin more subject to management quality, product cost, market monopoly power influence; Especially the operational cost is proportional to the net interest margin, average operating costs fell by 10%, net interest margin will be reduced by 43%. So during this period, operation cost is the most important one of the factors that affect the net interest margin.
Valverde & Fernández(2007)分析了包括德、法、英、意、荷、西班牙和瑞典等欧洲国家银行的数据，研究却发现相对于专业化银行，业务多样化的银行有更高的利差。与Maudos & Guevara(2004)一样，净利差与赫芬达尔—赫希曼指数(HHI)关系不显著，市场集中度与净利差没有必然的正相关关系，尤其是发现非传统业务收入可以“补偿”由于传统信贷业务竞争激烈而造成的利差下降。这个结果在一定程度上解释了欧洲银行市场垄断势力增加与净利差减少的矛盾现象(Cetorelli & Gambera, 2002)。因此政策制定者在分析银行市场垄断势力时应当考虑银行业务的多样化因素，而且对利差衡量不能仅局限于利息收入，应该使用更宽泛的银行利差衡量。Hawtrey & Liang(2008)在考察了14个OECD国家的银行1987－2001年的数据后，认为工业化国家银行净利差与市场垄断势力、运营成本、风险规避、利率风险、信贷风险、隐含利息支付等正相关，而与管理效率负相关。
Mr Valverde & ms Fernandez (2007) analyzed including Germany, France, Britain, Italy and other European countries, the Netherlands, Spain and Sweden bank data, the study found that relative to the specialized Banks, Banks have higher spreads business diversification. With Maudos & Guevara (2004), the net interest margin and fanta, -- Hector seaman index (HHI) no significant relations, market concentration is not necessarily positive correlation with net interest margin, especially can find non-traditional business income "compensation" due to the traditional credit business competitive spreads down. The results to a certain extent, explains the European banking market monopoly power increase with net interest margin decrease paradox (Cetorelli & Gambera, 2002). Therefore policy makers should consider when analyzing bank market monopoly power bank business of diversified factors, but also for spreads measure can not only limited to the interest income, larger Banks spreads measure should be used. Hawtrey & Liang (2008) investigated in 14 OECD countries after the bank data from 1987-2001, think that industrialized countries bank net interest margin and market monopoly power, operating cost, risk, interest rate risk, credit risk, such as implicit interest payments are related, and negatively related to the efficiency of management.
(二)对转型国家和地区银行的研究
(2) the study of transformation of national and regional Banks
与欧美发达国家不同，鉴于中东欧转型国家从20世纪90年代开始进行了利率市场化、银行私有化进程，银行体系发生了根本性变革。Drakos(2003)用GLS方法来评估中东欧(CEECs)与前苏联国家(FSU)83家银行转型10年的银行效率。在考虑了流动性风险、信用风险、利率风险、资本充足率之后，研究发现转型过程中银行净利差在减少，说明银行效率在提高，在一定程度上也表明了这些国家的银行市场化改革富有成效，尤其是外资银行的进入对提升银行系统效率和银行竞争力具有促进作用。
Unlike Europe and the United States developed countries, in view of the central and eastern European transition countries beginning in the 1990 s to the marketization of interest rates, bank privatisations, fundamental changes have taken place in the banking system. Drakos (2003) with GLS method to assess central and eastern Europe and the former Soviet union countries (CEECs) 10 years (FSU) 83 Banks transformation efficiency of Banks. Considering the liquidity risk, credit risk, interest rate risk and capital adequacy ratio, found the bank net interest margin in the transformation process in reducing, instructions in improving bank efficiency to a certain extent, also shows that the productive banking marketization reform in these countries, especially foreign Banks enter to promote the competitiveness of the banking system and bank efficiency has a promoting effect.
Schwaiger & Liebeg(2007)的研究认为，在CEEC国家中普遍存在着道德风险行为，信贷风险比利率风险在银行利差决定中更为重要，银行净利差减少主要是由于运营成本降低、效率提高和银行非传统业务的发展，整体而言，银行利差将随着金融深化和经济的发展而呈现下降趋势。但与Drakos(2003)研究结论不同，他们的研究中却发现在转型国家中外资银行的净利差更高，认为这主要是与外资银行较低的再融资成本相关。更有意思的是，他们的研究还发现，银行是否国有对银行利差并没有显著影响，因为随着市场化改革和经济的发展，国有银行在市场中的重要性下降，在激烈的市场竞争中，国有银行会倾向于模仿私人银行行为以提高其竞争力。⑥他们的研究结论与Micco et al.(2007)的相似，即产权的重要性只在转型国家的银行中体现，但在工业化国家中却得不到支持。这样的研究结论似乎表明在经济转型初期，银行的产权结构对银行绩效有着重要影响，但随着市场化改革的深入、国有经济比重的下降以及经济的快速发展，在市场化机制作用下，国有银行将不再扮演政策性支持的角色，此时产权的重要性也将下降。
Schwaiger & Liebeg (2007) research suggests that there is a widespread moral hazard behaviors in CEEC country, credit risk is more important than in the determination of interest rate risk spreads in the bank, the bank's net interest margin decrease is mainly due to reduce operating costs, improve efficiency and the development of Banks non-traditional business, overall, the bank spreads will appear with the development of financial deepening and economic decline. But with Drakos (2003) research conclusion is different, their study found that in the transition countries provided higher net interest margins, this was thought to be mainly associated with the refinancing costs of foreign Banks is relatively low. More interesting is that their study also found that whether the state-owned bank to bank spreads did not significantly influence, because with the development of the market-oriented reform and economy, the importance of state-owned Banks in the market decline, in the fierce market competition, state-owned Banks tend to imitate the private bank behaviour in order to improve its competitiveness. 6. Their research conclusion and Micco et al. (2007), namely, the importance of property rights only embodied in the transition countries' Banks, but in the industrialized countries. Research conclusion suggests that at the beginning of the economic transformation, the bank's property right structure has important effect on bank performance, but with the deepening of market-oriented reform, declining proportion of state-owned economy and the rapid development of economy, under the effect of market mechanism, the state-owned Banks will no longer play the role of policy support, at this point the importance of property rights will also decline.
在考虑银行自身因素(资本充足率、存款准备金率)、宏观经济环境(通货膨胀率、实际经济增长、实际短期利率)和银行产权异质性后，Claeys & Vennet(2008)对31个CEEC国家1994－2001年数据研究后认为，相对于西欧国家，CEEC国家银行有着更高的净利差，原因在于银行的低效率和银行市场的竞争条件。但随着转型国家银行信贷快速增长和外资银行的不断进入，市场结构对净利差的影响在减弱。
Considering the Banks themselves factors (capital adequacy ratio, reserve requirement ratio), macro economic environment (the rate of inflation, real economic growth, real short-term interest rates) and after the bank property heterogeneity, Claeys & Vennet (2008) 1994-2001 data of 31 CEEC countries after study that compared to Western Europe, CEEC national bank has a higher net interest margin, because of the low efficiency caused by the bank and the bank market competition conditions. But with the rapid growth of bank credit transition countries and foreign Banks continue to enter, in down market structure's impact on net interest margin.
加入产权因素后，对俄罗斯银行的研究认为，传统上决定银行利差的因素会有所差异，信用风险只对国内私有银行影响显著；规模因素对国内私人银行和外资银行有影响；市场集中度对国有和私人银行的影响不显著(这与发达国家不同)，但对外资银行显著；运营成本与风险规避对所有银行都有明显影响。
After joining property factors, the study of Russian Banks believe that traditionally decided to bank spreads will differ, the credit risk is only significant effects on domestic private Banks; Scale factors to influence domestic private Banks and foreign Banks; Influence of market concentration of state-owned and private Banks was not significant (different) with that of developed countries, but significant to foreign Banks; Operating costs and risk aversion have obvious effects on all Banks.
一项更加全面的研究是Kasman，Tunc & Vardar，Okan(2010)使用了从1995－2006年欧盟原有成员国、12个新加入的成员国以及3个候选国共29个国家1431家商业银行数据，并以2000年为界限分两个阶段进行研究，发现与前一阶段有所不同，在2000－2006年间资本充足率与净利差负相关，而且宏观变量对利差的影响也不显著；在对新旧成员国进行对比分析后发现，由于法律、金融和宏观环境不同，影响银行净利差的决定因素也不尽一致，原先成员国的银行利差主要受银行层面相关数据的显著影响，而新加入的成员国银行利差却主要受宏观层面数据的影响。但随着欧盟统一的进程，这种在宏微观不同层面影响的差异可能会消失。
Is a more comprehensive study Kasman, Tunc & Vardar, Okan (2010) from 1995-2006 years old eu members, 12 new members and three candidate countries 29 countries 1431 commercial bank data, and in 2000, for the boundary in two stages, found that is different from the previous phase, from 2000 to 2006, capital adequacy ratio is negatively related to the net interest margin, and macroeconomic variables affect spreads also not significant; Found after analyzing comparing old and new member states, due to legal, financial, and macro environment is different, have taken different factors affecting the decision of the bank's net interest margin, the original members of the bank spread mainly by bank level significantly affect the related data, and the new member bank spreads is mainly influenced by macro level data. But as the eu's unified, in the macro to microcosmic influence different levels differences may disappear.
(三)对拉美地区国家银行的研究
(3) the study of Latin America national bank
从20世纪70年代开始，拉美地区逐渐开始经济转型，解除金融抑制，随着利率市场化进程的导入，净利差问题也受到关注。Brock & Rojas(2000)针对Ho等提出的理论模型，对拉美地区的阿根廷、玻利维亚、哥伦比亚、智利、秘鲁等国90年代的商业银行进行了实证检验，部分结果虽然与欧美国家一致，运营成本、存款准备金要求与净利差正相关，但在拉美地区不良贷款率并不与净利差正相关，资本充足率与净利差也没有必然的联系。在不同国家，微观因素和宏观因素在净利差决定中的重要性也不相同。因此，不能简单的把工业化国家的模型套用在拉美地区。
Starting in the 1970 s, Latin America gradually began to economic transformation, remove financial repression, as the interest rate marketization process of import, the problem of net interest margin is also concerned. Brock & Rojas for Ho (2000) put forward theoretical model, the region of Argentina, Bolivia, Colombia, Chile, Peru and other countries in the 90 s of the commercial bank has carried on the empirical test, although some results are consistent with the European and American countries, operating costs, reserve requirements, and positively related to the net interest margin, but in Latin America non-performing loan ratio is not positively related to the net interest margin, capital adequacy and net interest margin is not necessarily linked. In different countries, the micro and macro factors in the importance of net interest margin decision is not the same. Therefore, cannot simple models to the industrialized countries in the region.
鉴于外资银行规模在拉美地区的显著增加，Martinez Peria & Mody(2004)研究了20世纪90年代中后期外资商业银行的参与和集中度对拉美的阿根廷、智利、秘鲁、哥伦比亚和墨西哥等国银行利差的影响。结果表明，从长期来看，外资银行的进入可以使该国银行体系运营成本降低，进而影响净利差；而且外资银行的进入主要是通过收购而非设立新的机构，使得银行集中度上升和竞争程度下降，这对运营成本和净利差都有着显著影响。
Given the scale of foreign Banks in the region increased significantly, Martinez Peria & Mody (2004) studied the mid-to-late 1990 s foreign commercial Banks' participation and concentration in Latin America in Argentina, Chile, Peru, Colombia and Mexico bank spreads. Results show that in the long run, foreign bank entry can reduce operating costs for the country's banking system, and then affect the net interest margin; And foreign Banks to enter mainly through acquisitions rather than set up a new agency, that bank concentration increased and the degree of competition, which have a significant impact on operating costs and net interest margin.
Gelos(2009)通过对85个发展中国家或新兴市场国家及地区(包括14个拉美国家)2200多家银行1999－2002年的数据分析研究后发现，通货膨胀、银行收益税、借款者信息、银行规模和资本充足率等是决定这些国家银行利差的共同因素，但相对于新兴国家，拉丁美洲商业银行因缺乏效率、利率水平和存款准备金过高、竞争程度差等因素而使其银行净利差更高。
Gelos (2009) in 85 developing countries or emerging market countries and regions, including 14 Latin American countries more than 2200 Banks in 1999-2002 data analysis found that inflation, bank information, bank size and gains, borrowers such as capital adequacy ratio is a common factor of these state bank spreads, but relative to the emerging countries, due to lack of efficiency of commercial Banks in Latin America, high interest rates and reserve requirements, poor competition factors and make the bank's net interest margin is higher.
(四)对亚洲地区国家银行的研究
(4) the study of state bank in Asia
Kannan, Narain & Ghosh(2001)对印度86家银行(27家公共银行、31家私人银行、28家外资银行)的研究表明，银行规模与净利差没有关系，中间业务发展有利于净利差提升，资本在公共银行的净利差中起重要作用，不良贷款对所有类型银行净利差都有重要影响。而Doliente(2003)在对东南亚国家(印度尼西亚、马来西亚、菲律宾和泰国等)的研究中却发现，这些国家比同期西欧国家银行净利差高个百分点，贷款质量、流动性资产、抵押品、运营成本、资本充足率、利率波动和市场结构等是影响银行净利差的主要因素。
Kannan, Narain & Ghosh (2001) on 86 Banks in India (27 public Banks, 31 private Banks, 28 foreign Banks) research has shown that bank size has nothing to do with net interest margin, the intermediary business development is conducive to net interest margin, capital plays an important role in public bank's net interest margin, for all types of bank non-performing loans have important influence on net interest margins. And Doliente (2003) in the southeast Asian countries (Indonesia, Malaysia, the Philippines and Thailand, etc.) study found that these countries of Western Europe over the same period bank net interest margin is % higher, the loan quality, liquid assets, collateral, operating costs, capital adequacy ratio, interest rate volatility and market structure are the main factors influencing the bank's net interest margin.
(五)全球银行净利差的比较研究
(5) a comparative study of the global bank net interest margin
Demirgiiu－Kunt & Huiainga(1999)用更为广泛的跨国间数据(包括80个国家1988－1995年数据)，在分析银行自身特征和银行周边环境对利差的影响后认为，发展中国家外资银行相对于国内银行而言有着较高的净利差和利润，宏观经济条件如通货膨胀与银行净利差正相关，高利率意味着高利差和利润，这一切在发达国家则正好相反。就金融结构而言，银行集中度与银行利差正相关，股票市场的发展则会降低银行净利差；制度因素如信贷权利(Credit Right)、法律规则、破产和准备金等在发展中国家对净利差和利润的影响比其在发达国家对净利差和利润的影响大。
Demirgiiu - Kunt & Huiainga (1999) with a broader cross-border data (including data in 1988-1995) in 80 countries, the analysis of characteristics of the Banks themselves and peripheral environment influence to spread after the bank believes that foreign Banks in developing countries relative to the domestic Banks have higher net interest margin and profitability, macroeconomic conditions such as the inflation is positively related to the bank's net interest margin and higher interest rates mean high spreads and profits, all this is just the opposite. In developed countries In terms of financial structure, bank concentration and bank spreads positive correlation, the development of the stock market will reduce the bank's net interest margin; Institutional factors such as Credit rights (Credit Right), the rule of law, bankruptcy and reserves in developing countries such as impact on net interest margins and profits than its impact on net interest margins and profits in developed countries.
Demirgiiu－Kunt，Laeven & Levine(2003)进一步用1995－1999年72个国家1400多家银行的数据分析了银行规则、集中度和制度对银行利差的影响。研究发现，严格的银行规则(如对银行进入和银行活动的限制)可以提高银行净利差；但在考虑了国家制度(如经济自由度、产权保护等)后，这种作用却不明显。银行业集中度与净利差的关系更加复杂，就银行层面而言，行业集中度与净利差显著正相关；但是加入规则约束和宏观条件后，这种关系在减弱。总体而言，银行净利差受银行自身因素(如规模、杠杆率、商业类型、产权结构等)、银行市场结构、宏观经济条件如通货膨胀、税收和监管、金融结构以及法制等制度因素影响。
Demirgiiu - Kunt Laeven & Levine (2003) further in 1995-1999, more than 1400 Banks in 72 countries data analysis system of the bank rules, concentration and influence on bank spreads. Studies have found that strict rules of the bank (such as restrictions on entry and bank activity) can improve the bank's net interest margin; But considering the national institution (., economic freedom, property rights protection, etc.), the effect is not obvious. Centralized banking relationship with net interest margin is more complicated, in terms of bank level, industry concentration ratio and net interest margin was significantly positive correlation; But after adding rules constraint and macro conditions, in down this relationship. Overall, the bank's net interest margin by the Banks themselves factors (such as size, leverage ratio, type of business, property right structure, etc.), the structure of the banking market, macroeconomic conditions, such as inflation, taxes and regulation, financial structure and legal system and other system factors.
在Demirgiiu－Kunt(1999，2003)研究的基础上，Dietrich et al.(2010)用1994－2008年96个国家的数据，重点考察国家层面的宏观治理因素对银行净利差的影响，根据经济发展水平不同，把样本分为发达国家和发展中国家两组，发现决定发达国家和发展中国家银行利差因素有着显著差异，其中治理因素(债权指数、执行效率、合约执行时间天数、法律规则、信息分享机制等)对发展中国家银行利差的影响更为显著；相对而言，银行层面因素(运营成本、贷款损失率、存款准备金要求、行业集中度等)对发达国家银行利差的影响更大。
In Demirgiiu - Kunt (1999200 3), on the basis of research, Dietrich et al. (2010) 1994-2008 data of 96 countries, the macroscopical management that focuses on country level factors influence on bank's net interest margin, according to the economic development level is different, the sample is divided into two groups in both developed and developing countries, found that developed countries and developing countries decided to bank spreads factors have significant differences, including governance factors (index, execution efficiency, creditor's rights contract execution time of days and the rule of law, information share mechanism, etc.) affect developing countries bank spreads more pronounced; , in contrast, bank level factors (operating costs, loan loss rate, reserve requirements, industry concentration, etc.) a greater influence on Banks in the developed countries spread.
(六)对国内银行净利差决定因素的研究
(6) study on determinants of domestic bank net interest margin
长期以来，我国银行利率存在着较为严格的管制，利差问题并没有引起学界重视，但随着上世纪90年代以来商业银行利率市场化改革的推进，对净利差决定因素的研究也开始关注。黄国平等(2007)在构建的理论模型中认为，存贷款利差的合理水平取决于违约率、损失率、流动性风险、资本缓冲与信贷类资产的比率，以及无风险基准利率。周开国等(2008)以Ho(1981)提出的交易者模型为基础，通过引入平均运营成本、资产规模和其他因市场不完美引起的因素，对中国商业银行的净利差决定因素进行实证检验后发现，市场竞争结构是商业银行净利差的重要决定因素之一；平均运营成本、交易规模、隐含利息支付、准备金的机会成本与净利差正相关；而管理效率、资产规模、风险厌恶程度、信用风险与净利差成反比。
Exist for a long time, China's interest rates relatively strict control and carry problem and not to the attention of the academia, but as the commercial bank interest rate marketization reform since 1990 s, the net interest margin determinant research also begin to pay close attention to. Yellow equality (2007) argues that in building a theoretical model reasonable margin level will depend on the default rate, loss rates, liquidity risk, the ratio of capital cushion to credit assets, as well as the risk-free benchmark interest rates. Zhou Kaiguo etc. (2008) to Ho (1981) model proposed by traders as the foundation, by introducing the average operating costs, asset size and other factors caused by market imperfections, the net interest margin of China's commercial Banks after the determinants empirically test found that competition in the market structure is one of the important determinants of commercial bank net interest margin; Average operating costs and transaction scale, implicit interest payments, reserves of opportunity cost is positively related to the net interest margin; The efficiency of management, asset scale, the degree of risk aversion, credit risk is inversely proportional to the net interest margin.
周鸿卫等(2008)利用中国26家商业银行1999－2006年数据进行分阶段实证研究，结果表明，从1999－2004年间，对我国商业银行净利差影响显著的因素有存贷基准利差、银行资本充足性、经营成本、储备的机会成本；2005－2006年间影响我国商业银行净利差比较显著的因素有违约风险、资本充足性、经营成本、储备的机会成本、资产管理质量、流动性风险等。这说明我国商业银行的价格行为不仅受基准利率的影响，而且与各种风险因素和资产负债管理有关，我国商业银行存贷款自主定价的能力正在逐步形成。
Zhou Hongwei etc. 26 (2008) use of China's commercial Banks in 1999-2006 data for empirical research in stages, the results show that from 1999-2004, to our country commercial bank net interest margins significantly influence factors have benchmark savings and loan spreads, bank capital adequacy, operation cost, the opportunity cost of reserves; Affect our country commercial bank net interest margin between 2005 and 2006 more significant factors include default risk, capital adequacy, opportunity costs, operating costs, reserve asset management quality, liquidity risk, etc. This shows the price behavior of Chinese commercial Banks not only affected by the benchmark interest rate, and with a variety of risk factors and related assets and liabilities management, our country commercial bank lending and deposit are gradually formed independent pricing ability.
考虑到市场垄断势力及中间业务等因素对银行利差的影响，赵旭(2009)借鉴Valverde & Fernandez(2007)的方法，估算了中国15家主要商业银行1998－2006年的存贷利差、会计利差和边际利差，研究结果表明，商业银行的实际存贷利差简单平均为％，与发达国家相比，我国银行利差基本处于一个适中且偏高的水平，且呈上升态势。在对影响银行利差的主要因素分析中发现，商业银行的风险管理和中间业务发展对银行利差影响不显著。国有商业银行机会成本与银行利差存在显著的正相关关系，宏观经济变量对银行利差有一定的影响，但影响并不显著，而反映市场竞争水平的市场结构变量对银行利差的影响却不容忽视。
Considering the market monopoly power and intermediary business to bank spreads, the influence of the profit-structure (2009) from Mr Valverde & ms Fernandez (2007), the method of estimating the 15 major commercial bank of China in 1998-2006 spreads and marginal spreads spreads and accounting, the research results show that the actual loan of commercial Banks carry simple average is %, compared with the developed countries, our country basic banking spreads in a moderate and high levels, and on the rise. In the analysis of the main factors influencing the bank spreads, found that the risk management of commercial Banks and spreads on bank middle business development effect was not significant. Opportunity cost and the state-owned commercial bank spreads a significant positive correlation, macro economic variables have a certain influence on bank spreads, but the impact is not significant, and reflect the market structure of market competition level variables impact on bank spreads cannot be ignored.
在做市商模型的基础上，程茂勇、赵红(2010)运用我国14家上市商业银行1999－2008年的面板数据，从静态和动态两个方面对我国银行利差影响因素进行分析，发现市场垄断程度、运营成本、风险规避程度、利率风险与信用风险的交叉项、交易规模、中间业务的佣金及费用、运营效率等因素在各种情况下均与利差有显著的相关关系；动态回归分析显示上一期利差对本期利差也有显著的影响，其中市场垄断势力、运营成本对利差的影响最大。
The market-maker model, on the basis of Cheng Maoyong, Zhao Gong (2010) using the 14 listed commercial Banks in China in 1999-2008 panel data, from the static and dynamic two aspects to our country bank spreads influence factors were analyzed, and found that the degree of market monopoly, operating costs, the degree of risk aversion, interest rate risk and credit risk of cross terms, trading scale, the intermediary business of fees and costs, operating efficiency under various circumstances and carry a significant correlation; Dynamic of the regression analysis shows a period spreads spreads also have a significant impact, including the market monopoly power, the influence of the operating costs of spread is the largest.
四、文献简要评述与启示
Four, literature is briefly reviewed, and enlightenment
从已有文献可以看出，目前学术界对于净利差的计算尚无完全一致的方法，对于净利差决定因素的讨论大致可以分为三类：一是微观层面受到如银行规模、资产质量、运营成本、信贷产品、管理水平及其面临的利率风险、信用风险等影响；二是在行业层面上与市场结构、市场垄断势力及产权特征等因素相关；三是受经济增长、通货膨胀、法律制度等宏观经济层面影响。在不同的国家和地区，相同因素所起的正反作用并不一致，而且不同因素所起的重要程度也不尽相同，这可能与其经济发展水平密切相关。相对而言，由于发达国家制度基本稳定，因而微观基础环境对发达国家或地区商业银行的利差影响更大，而在广大发展中国家尤其是转型国家，由于制度转型，宏观制度环境的作用显得更为重要。甚至对同一国家或地区而言，经济发展阶段的不同，即使是相同的因素对银行净利差的作用效果也会有很大的差异，因此在研究中应当区别对待。
Can be seen from the existing literature, the current academia for calculation is completely consistent with the method of net interest margin, net interest margin for the discussion of the determinants can be roughly classified into three kinds: one is the size of the micro level by such as Banks, asset quality, operating costs, credit products, and the interest rate risk, credit risk management level, etc; Second, at the industry level and the market structure, market monopoly power and property right characteristic and other related factors; Three is influenced by economic growth, inflation and the economy as a whole legal system, etc. In different countries and regions, the same factors of positive and negative role, but also to the importance of the different factors that is not the same, it may be closely related with its economic development level. Relatively, due to the system of developed countries basically stable, and the micro base environment in developed countries or regions affected more by commercial bank interest rate differentials, and especially in developing countries and transition countries due to the system transformation, the role of macroscopic system environment is more important. Even for the same country or region, the different economic development stage, even if is the same the effects of factors on the bank's net interest margin will also have very big difference, so in the study should be treated differently.
基于西方发达国家稳定的经济制度背景，净利差作为金融中介机构成本的反映，利差水平及其波动性是商业银行经营管理的目标。然而我们也应当注意到，用净利差来衡量商业银行资产经营效率有其不足，应与其他指标共同使用。在不同国家或地区，宏微观因素在净利差决定中的重要性也不相同，因此不能简单地把工业化国家商业银行净利差模型简单套用在发展中国家尤其是转型国家中(Brock & Rojas，2000)，尤其是在中国银行业制度转型与利率尚存在一定管制的背景下，存贷利差水平仍难以完全反映出资金的真实供求状况。⑦虽然目前国内对商业银行净利差的研究更多的是对其决定因素展开实证分析，这对理解中国银行业的利差问题有促进意义，但在利率市场化渐进改革的过程中，中国银行业运行的基本经济制度正在发生根本性变革，我们对银行净利差水平及波动性的影响因素研究就不能仅满足于对西方国家已有银行利差模型的简单套用，而更应当注重中国银行业利差演进的内在制度逻辑及其在中国现实中所特有的经济内涵，这是今后研究的方向。对于货币政策中是否应当关注净利差波动的问题，目前学者们研究的也较少而且意见不一。不管怎样，我们发现在经济发展的不同周期中银行净利差水平存在很大差异，不可否认利差水平的这种波动会对银行产生不同的冲击效果，进而也对整体经济运行产生不同的影响，这个问题也有待更进一步的研究。
Based on western developed countries and stable economic system background, net interest margin as reflect the costs of the financial intermediaries, spreads is the goal of commercial bank management level and volatility. But we should also note that, with net interest margin to measure commercial bank assets management efficiency has its limits, should be used together with the other metrics. In different countries or regions, macro to micro factors in the importance of net interest margin decision also is not the same, so can't simply commercial bank net interest margin model built by the industrialized countries in the developing world, especially in the transition countries (Brock & Rojas, 2000), especially in the transformation of China's banking system and the interest rate is certain control of the background, the level of spreads are still hard to fully reflect the real supply and demand of funds. 7) although the present domestic research on commercial bank's net interest margin is more empirical analysis on its determinants, it spreads in understanding China's banking problem is promoting significance, but in the process of market-oriented interest rate reform gradually, China's banking operation of the basic economic system is undergoing fundamental change, our bank net interest margin level and the influence factors of volatility research can not only meet in the west bank spreads models available simply copy, but more should pay attention to China's banking sector spreads evolution system of the internal logic and its peculiar to the real economy connotation in China, this is the direction of future research. About whether monetary policy should focus on net interest margin fluctuation problem, the scholars study less and disagree. Anyway, we found that the bank's net interest margin in the different period of economic development level there is a big difference, there is no denying that spreads such volatility levels will have different impact on Banks, and also different impact on overall economic operation, the problem also needs to be further research.
注释：
Annotation:
①除了本文所介绍的净利差外，在银行利差中，国内学者更多的是使用毛利差的概念，即贷款利率减去存款利率，净利差与毛利差在一定条件下可以转换，国内学者李波(2005)区分了毛利差与净利差。
(1) in addition to the net interest margin, presented in this paper, in bank spreads, domestic scholars more is the difference in using margin concept, namely the loan interest rate minus the deposit rate, net interest margin and gross profit difference under certain conditions can be transformed, the domestic scholar li (2005) distinguishes between gross profit and net interest margin.
②其他的几种测算公式分别为(利息收入/贷款总额)-(利息支出/存款总额)、(利息收入/所有生息资产)-(利息支出/所有生息负债)、(利息收入+佣金收入)/贷款-(利息支出+佣金支出)/存款、(贷款利息收入/贷款)-(存款利息支出/存款)、(利息收入-利息支出)/总资产。
(2) other several calculation formulas respectively (interest income/total) - (interest payments/deposits), / all interest-bearing assets (interest income) - (interest payments/all interest-bearing liabilities), commission income (interest income +)/loan - (interest payments plus commission payments)/deposit, loan interest income/loan - (deposit payments/deposit), interest income - interest payments)/total assets.
③Zarruk(1989)的结论还发现，运营成本增加对银行利差影响无法确定，而且大型银行的利差将更高，这与Ho(1981)的经验研究所得出的结论完全不同。
The conclusion of (3) Zarruk (1989) also found that operating costs increase spreads influence can not be sure for the Banks, and big Banks spreads will be higher, with the Ho (1981) empirical research institute's conclusion is completely different.
④该文中的货币中心银行参照美国标普公司(1993)和Keefe(1993)等的标准，是指如花旗等29家在全球金融市场中具有重要地位的大型商业银行；跨区域性银行是根据美国商业周刊(1993年5月3日)的银行排名，是除了货币中心银行外的跨州经营的90家大型商业银行；区域性银行是总部设在区域金融中心，资产在100亿~250亿美元的银行；地方性银行是小型银行，如同社区银行，资产规模大约在10亿~100亿美元。
(4) in this paper, a money center bank reference (1993) and Keefe s&p companies in the United States (1993) standard, refers to such as citigroup, 29 in the global financial market has the important position in the big commercial Banks; Cross regional bank is according to the United States by business week (on May 3, 1993) the bank number, is apart from the money center Banks to operate across state of 90 large commercial Banks; Regional bank is headquartered in regional financial center, assets in 10 billion to $25 billion in bank; Is a small local bank as community Banks, assets at about $1 billion to $10 billion.
⑤Hanweck and Ryu(2004)利用1986第一季度到2003年第二季度的季度数据同样证实了这一关系，相对而言大银行更容易受信用冲击，受利率或期限结构冲击更弱。
5 Hanweck and Ryu (2004) used in the first quarter of 1986 to 2004 in the second quarter of the quarterly data confirmed the same relationship, relatively large Banks are more vulnerable to credit, interest rates or term structure impact weaker.
⑥Schwaiger and Liebeg(2007)认为，这与研究样本的时间段有关，在Drakos(2003)的研究中使用的样本数据为1993~1999年，此时CEEC国家正处于转型时期，而他们的研究样本数据是2000~2005年，这时期这些国家已经基本完成转型，经济快速发展，开始向工业化国家逐步迈进。
6 Schwaiger and Liebeg (2007) argues that it has to do with the sample period, in Drakos (2003), which is used to study sample data of 1993 ~ 1999, the CEEC country is in transition period, and their study sample data is 2000 ~ 2005, this period has been basically completed transformation, these countries rapid economic development, began to industrialized countries.
