1、 外文翻译 Banking supervision and nonperforming loans: a cross-country analysis Material Source: Journal of Financial Economic Policy VO l. 1 No. 4, 2009. Author: Abd-el-Kader Bourikas, Neil Bouillon Tactic and Sana Jealously DEFI, University of Tunis, Montclair, Tunisia Introduction After a relatively
2、calm decade, the international banking industry suffered during the last three years an unprecedented meltdown. Several banks throughout the world, including in developed and developing countries, experienced severe losses on their credit portfolios leading to banks failures and to a global fear of
3、a systemic crisis. This crisis raised further concerns about financial systems stability and the need for a closer control and supervision on lending activities and institutions. In particular, international regulators (IMF, World Bank, and the BIS) engaged, over the last decade, several reforms, an
4、d programs aiming to strengthen banking and financial systems in different countries. Various assessments are periodically made to timely predict any undesirable exposure. Specifically, the aggregate rate of nonperforming loans (NPL s) is commonly used as a soundness indicator. Bank industry factors
5、, supervision and NPL s: literature review and hypothesis development The empirical literature on NPL s explains banks problem loans using bank specific or internal variables. These bank-specific factors are related to bank management and firm-level features which include proxies of bank capitalizat
6、ion, provisioning policy, profitability, ownership status, and industry concentration. Bank capitalization. Theoretically, the capital adequacy ratio (CAR) might serve as a tool to control excessive risk taking by banks and to prevent them from being insolvent through recapitalization (Basel accord)
7、. Banks with CAR less than the regulatory minimum are forced to adjust their balance sheet to comply with there requirements either by raising more capital (holding assets constant) or reducing risk-weighted assets (holding capital constant). In fact, raising the level of capital relative to risky a
8、ssets by either means could have a beneficial impact on the bank performance and soundness (Fries et al., 2002). Indeed, Koehn and San tom e r o (1980) show theoretically that the portfolio risk increases with the increase of them in mum capital ratio. They state that banks under pressure to increas
9、e capital will reach the desired level by increasing the risk of assets. Capital adequacy ratio is negatively associated with NPL s. Provisioning policy. Loan loss provisions are regarded as a controlling mechanism over expected loan losses. Under backward-looking provisioning practices, where provi
10、sions are triggered by default incidents on loans, higher levels of NPL s are associated with high rates of pro-visioning (Hassan and Wall, 2004). At the same time, banks anticipating high levels of capital losses might create higher provisions to decrease earnings volatility and to reinforce medium
11、 term bank solvency. In this case, managers can also use loan loss provisions to signal the financial strength of their banks. The willingness of a bank to provision for loan losses is regarded as a strong belief in the future performance of the bank (Ahmad et al., 1999). The overall rate of provisi
12、oning reflects the general attitude of the banking system toward risk control. Bank profitability. Bank profitability may also determine the risk taking behavior of managers. Banks with high profitability are less pressured to revenue creation and thus less constrained to engage in risky credit offe
13、rings. At the same time, inefficient banks are more likely to experience high levels of problem loans. Poor management can imply weak monitoring for both operating costs and credit quality of customers, which will induce high levels of capital losses. Under this bad manage menthypothes is advanced b
14、y Berger and De Young (1997), managers lack competencies to effectively assess and control risks incurred when lending to new customers. Godless (2004), using the adjusted return on assets ratio (ROA) as a proxy for performance, shows that banks profitability negatively impacts the level of NPL s ra
15、tio. However, using a panel of 129 Spain banks during 1993-2000, Garcia- Marco and Robles-Fernandez (2007) find that higher levels of return on equity are followed by greater risk in the subsequent periods. They argue that profit-maximizing policies will be accompanied by higher levels of risk. Owne
16、rship status. The ownership status of the bank is associated with NPL s. Several studies document that State ownership may explain the behavior of risk taking of bankers and consequently the level of NPL s. For instance, Salas and Saurian(2002) argue that to enhance the economic development of the c
17、ountry, state-owned banks have more incentives to fund riskier projects and to allocate more favorable credits for small and medium firms. This inadequate risk taking behavior (compared to the return profile) will lead to a higher level of NPL s. In the same vein, Microtel. (2004) report that state-
18、owned banks tend to have higher levels of NPL s, due to their weak credit recovery capacity compared to privately owned banks. Others suggest that the interaction between private and state shareholding in the same bank could determine the risk level taken by banks. Heal. (2004) argue that unjustifie
19、d risky behavior is lower when the two groups check and balance each other. In the opposite, when private and state shareholders collude, especially in societies with little civil disciplines, problem loans will be higher due to risky credit offering. Tina (2000) suggests that under conditions of ma
20、rket imperfection, due to a balancing mechanism between management incentives and bureaucracy forces, a mixed enterprise (joint shareholding of private and state owners) will maximize social surplus. Empirically, Novas and Welling (1995) report lower performance for state controlled banks in Brazil
21、and Argentina due to high proportion of problem loans given to government. Mice et al. (2004), analyze 50,000 financial institutions with different ownership types covering 119 countries. They conclude that NPL s tends to be higher for banks with state ownership than for other groups. This is explai
22、ned by the development mandate given to state-owned banks in developing economies. Hub et al. (2004) use apneal of Taiwanese banks and find a positive correlation between capital share owned by the state and the level of NPL s. Finally, Garcia-Marco and Robles-Fernandez (2007) investigate the relati
23、onship between risk taking and ownership structure. They document that commercial banks (mainly private owned) are more exposed to risk than deposit banks (mainly state owned). Industry concentration is associated with NPL s. By its nature, the banking sector should be regulated and supervised to en
24、sure the stability of the whole financial system. During the recent decades, the banking regulatory framework has experienced sharp changes. Several reforms regarding banking supervision have been initiated since 1988 (Basel I) and reviewed since then until 2004 (Basel II). The question of how regul
25、ation influences the banking stability and soundness remains a source of debate. In previous studies, there is no consensus on what type of regulations and supervisory practices promote bank development, enhance financial stability, and facilitate efficient corporate finance (Barth et al., 2004; Bec
26、k et al., 2006b; Shaffer, 2008). In the remainder of this section, we examine the impact of the regulatory framework on problem loans. We use four variables related to the level of capital requirement, the official supervisory power, the market discipline, and the independence of supervisory authori
27、ty. Capital stringency. Regulatory and supervisory bodies emphasize the positive role of capital stringency as a buffer against losses and hence failures (De warty pentad Tirol, 1994). However, empirical evidence suggests that this is not always the case. Barth et al. (2004) study the relationship b
28、etween specific regulatory and supervisory practices and banking-sector development, efficiency and fragility. They find that stringent capital requirements are associated with fewer NPL s but are not robustly linked to other banking outcomes. Pastorals (2008) report a positive association between t
29、echnical efficiency and capital requirements, albeit not statistically significant in all cases. Other studies indicate that capital requirement increase, on the contrary, risk-taking behavior (Bosanko and Kanatas, 1996; Blum, 1999). For instance, Godless (2004), reports that capital regulation in t
30、he banking industry is positively related to excess risk taking. This increased credit risk leads to an increase in the ratio of NPL s. He explains that stringent constraints on capital imply additional pressure on assets returns, which could be done through higher risk taking. Pastors et al. (2006)
31、 find a negative relation-ship between capital requirements and banks soundness as measured by Fitch ratings. Finally, Delis et al. (2008) examining apneal of transition countries, argue that capital requirements do not have a statistically significant impact on productivity. Empirical results Table
32、 V shows the empirical results of our regressions. Model 1 presents results for the baseline model. Models 2 to 5 exhibit results for the supervision factors. The estimated coefficients on the banking industry variables appear to be robust to the specification used. The regressions show evidence for
33、 a negative impact of the variable (Dicer) on the credit risk exposure. This indicates that the regulatory capital serves as an indicator of the financial risk exposure of the whole banking system. These results suggest that the CAR might be used as a regulatory device to mitigate banks excessive ri
34、sk taking (Fries et al., 2002; Sinker and Greenwell, 1991). In conclusion, it appears that the first pillar of Basel II does not have any statistically significant impact on NPL s. The estimated coefficient for the lagged growth rate is not significant, indicating that economic conditions are not re
35、lated to problem loans. This result is contradictory to both theoretical prediction and common wisdom, which relate economic cycle to credit quality of borrowers (Sinker and Greenwell, 1991; Salas and Saurian, 2002). However, we document that the level of problem loans is higher in less financially
36、develop economies. Conclusion In this paper, we proposed an empirical framework to investigate the bank industry factors and supervisory determinants of NPL s on a cross-country basis. First, we use specific variables that capture many of the factors suggested by the theory and highlighted by case s
37、tudies. Besides, to investigate the role of the regulatory framework on credit risk outcomes, we introduce variables on banks supervision. Finally, to assess the impact of the effective implementation of those regulations, we experiment interactions of three institutional variables (corruption, demo
38、cracy, and rule of law) with each of the supervision proxies. In contrast with previous work, we include interaction terms between political and business environment variables and each of the supervision variables to investigate their impact on banks credit exposures. However, our results suffer fro
39、m the fact that the measures used only relate to statutory powers. Thus, they do not address the issue of the effective implementation of supervisory reforms. To investigate this channel, we introduce three interactions using the level of corruption, the degree of political openness, and the rule of
40、 law. All of these variables are supposed to have an impact on the efficacy of regulation. Our findings do not support the view that market discipline (albeit not strongly) leads to better economic outcomes and to reduce the level of problem loans. Our contention is drawn upon the absence of any ass
41、ociation between the variab le private monitoring and the level of problem loans. Indeed, using various specifications (interactions terms and sub samples of developed and developing countries) the coefficients never entered significantly. Our results have the following policy implications. First, h
42、igher Cars results in less credit exposures. Second, more developed financial systems experience improved stability. Hence, international regulators should continue their efforts to enhance financial development. Third, private ownership and foreign participation insure healthier financial systems i
43、n less developing economies. Although we do not show evidence for causality, our results suggest that foreign participation plays an important role in reducing credit exposure of financial institutions. However, in developed countries foreign entry led to more problem loans. Finally, to reduce credi
44、t risk exposure in countries with weak institutions, the effective way to do it is through enhancing the legal system, strengthening institutions, and increasing transparency and democracy. 银行监管与银行不良贷款：一种跨国家分析 资料来源： 财政经济政策册期刊 J. 2009(4)，第 1 期， p286- 318 作者：阿卜杜勒埃尔 卡迪尔 ， 尼尔布永和萨那基度 DEFI， 突尼斯大学，蒙特克莱，突尼斯
45、， 引 言 经过相对平静 的十年 ，国际银行业在过去三年遭受了前所未有的危机。几家银行在世界各地，在发达国家和发展中国家，包括经验丰富的领导对银行倒闭和一个系统性危机的全球投资组合的的信用严重损失 、 恐惧 加深。 这一危机提出了关于金融体 系稳定，为更紧密的控制和监督机构的贷款活动和需要进一步关注 的观点。特别是 （国际货币基金组织，世界银行和国际清算银行）从事国际监管机构 ， 在过去十年中，一些旨在加强和方案 的改革，在不同国家的银行和金融体系 各种评估将定期，及时 进行 预测。具体来说，不良贷款总额率常用的一种健全的指标。 银行行业因素，监督和不良贷款：文献回顾与假设发展 对银行不良贷款
46、的实证文献说明使用银行贷款的具体问题或内部变量。这些银行特有因素与银行管理和企业级功能，其中包括银行资本代理，供应政策，盈利能力，所有权状况以及行业集中度。 银 行资本 。 从 理论上讲，资本充足比率（ CAR）可能作为一种工具来控制由银行承担 的 过度的风险，并防止通过资产重组（巴塞尔协议）资不抵债 的状况 。与 CAR 低于法定最低要求银行被迫调整筹集更多资本（控股资产常数）或减少风险加权资产（控股资本常数）资产负债表或者有符合要求 的情况 。事实上，无论是指通过提高资本水平 、 相对风险资产 ，都 可能对银行的业绩和稳健性（ Friesetal.， 2002 年） 造成 有利 的 影响。
47、事实上，科恩和圣汤姆主任（ 1980年 ）认为 从理论上说，随着比率 的 增加投资组合的风险增加。他们指出，根据银行增加资本的压力 与 增加资产 的风险所需的水平 紧密相关 。 资本充足比率是负相关的不良贷款 拨备政策 。 贷款损失准备金被视为预期的贷款损失控制机制。从向后看的配置做法看，其中规定由贷款违约事件引发的，较高的不良贷款与（ Hassan 和长城， 2004 年）高利率水平相关。与此同时，银行的资本损失预测可能创造更高的高度规定，减少收入的波动，并加强中期银行的偿付能力。在这种情况下，管理者也可以使用贷款损失准备金以表示他们的银行财务实力。一个为银行贷款损失准备的意愿被视为在该银行
48、（ Ahmadetal.， 1999 年）未来业绩的坚强信念。 总的准 备金率反映了 银行对风险控制体系的总体态度。 银行盈利能力 。 银行盈利能力可能也决定了管理者的冒险行为。具有高盈利的银行，对于收入少的压力，从而少被迫去创造、从事高风险的信贷产品。在同一时间，效率低下的银行更可能经历的高问题贷款的水平。管理不善可能意味着无法经营成本和为客户提供优质的信贷，这将导致资本损失、高水平监测不力。在这种恶劣的管理是由 Berger 和代扬（ 1997 年 ）先进提出的先进的思维，由于管理人员缺乏能力，如何进行有效地评估和控制借贷时所产生新客户的风险。无神（ 2004 年 ），使用的资产的比例作为
49、绩效回报，经过代理 （办事处）调整后，结果表明，银行的盈利能力产生负面影响的不良贷款比率水平。但是，在 1993-2000 年的使用期间，加西亚， Marco 和罗夫莱斯 费尔南德斯（ 2007 年）发现，较高的股本回报水平是在随后的时期更大的风险。这是在对西班牙银行的 129 个面板分析之后，他们认为，利润最大化的政策会受到风险较高的陪同。 所有权状况。该银行拥有的地位是与不良贷款几项国家所有权的研究文件可以解释的危险行为，因此银行采取了不良贷款的水平。例如，萨拉斯和蜥蜴（ 2002 年 ）认为，为加强该国的经济发展，国有银行有更多的诱因，风 险较高的项目和资金分配为小型和中小型公司更有利的信贷。这不足的冒险行为（相对于回报状况）会导致不良贷款更上高的层次。同样的 Microtel（ 2004 年 ）报告说，国有银行与私有银行相比 ,往往有较高水平的不良贷款，这是由于私有银行他们的信贷的恢复能力弱的。也有人认为，私营和国有股之间在同一银行的相互作用可以决定银行采取的风险水平。 Heal （ 2004 年 ）认为，降低不合理的危险行为时，两组相互制衡。相反的，当私营和国有股东勾结，特别是在几乎没有民间社会学科，问题贷款将因更高的信贷风险的产品而随之产生。蒂娜（ 2000 年 ）提出，根据市场不完善的现象，由于管理之间的激励和官僚势
