内部控制中英文文献

Disclosure on Internal Control Systems As a Substitute of Alternative Governance

Mechanisms

According to agency theory, various governance mechanisms reduce the agency problem between investors and management (Jensen and Meckling， 1976； Gillan, 2006)。 Traditionally， governance mechanisms have been identified as internal or external。 Internal mechanisms include the board of directors, its role， structure and composition (Fama， 1980； Fama and Jensen, 1983）， managerial share ownership （Jensen and Meckling, 1976） and incentives， the supervisory role played by large shareholders （Demsetz and Lehn， 1985）， the internal control system (Bushman and Smith， 2001)， bylaw and charter provisions （anti-takeover measures） and the use of debt financing (Jensen， 1993）。 External control is exerted by the market for corporate control （Grossman and Hart， 1980）， the managerial labor market （Fama, 1980) and the product market （Hart， 1983).

After the various financial scandals that have shaken investors worldwide, corporate governance best practices have stressed in particular the key role played by the internal control system (ICS） in the governance of the firm。 Internal control systems contribute to the protection of investors’ interests both by promoting and giving assurance on the reliability of financial reporting， and by addressing the boards’ attention on the timely identification， evaluation and management of risks that may compromise the attainment of corporate goals. These functions have been widely recognized by the most diffused frameworks for the design of ICS that have stated the centrality of internal control systems in providing reasonable assurance to investors regarding the achievement of objectives concerning the effectiveness and efficiency of operations， the reliability of financial reporting and the compliance with laws and regulations （COSO， 1992; 2004).

Notwithstanding their relevance, investors cannot directly observe ICSs and therefore cannot get information on their design and functioning because they are

internal mechanisms， activities and processes put in place within the organization （Deumes and Knechel， 2008）。

As investors take into account the costs they sustain to monitor management when pricing their claims (Jensen and Meckling 1976）, management have incentives to communicate information on the characteristics of the ICS in order to inform investors on the effectiveness of ICS when other monitoring mechanisms （the ownership structure of the firm and the board of directors) are weak， and thereby providing them with the convenient level of monitoring (Leftwich et al。, 1981）。 The possible existence of substitution among different mechanisms has been debated in corporate governance literature （Rediker and Seth, 1995； Fernandez and Arrondo， 2005） based on Williamson’s （1983) substitute hypothesis, which argues that the marginal role of a particular control mechanism depends upon its relative importance in the governance system of the firm。

In this paper， we contend that disclosure on the characteristics of ICS is a relevant alternative governance mechanism in the monitoring package selected by the management. According to Leftwich et al. (1981) “managers select a monitoring package， and the composition of the chosen package depends on the costs and benefits of the various monitoring devices” (p. 59)。

In particular， we focus particular on the relationship between ICS disclosure and two other mechanisms of the monitoring package （ the ownership structure of the firm and the board of directors) that according to literature (Jensen and Meckling， 1976； Fernandez and Arrondo,2005; Gillan, 2006） play a relevant role in monitoring management's behavior。 We posit that incentives for reporting on the characteristics of ICS depend on the supervisory role played by the firms’ ownership structure and board of directors.

We therefore examine the contents and extent of ICS disclosure of 160 European firms listed in four different stock exchanges （London， Paris， Frankfurt and Milan） on a three—year period （2003 – 2005)。 By using this international sample， we are able to the depict some features of different institutional environments. We find evidence that disclosure on ICS is a substitute for the monitoring role

played by other governance mechanisms as ownership concentration， institutional ownership, the proportion of independent directors sitting on the board and the proportion of accounting expert members on the audit committee。

We add to previous literature on the governance role played by disclosure on ICS by adopting a complete disclosure framework that allows us to consider in detail the content and extent of information the management discretionarily communicates on the ICS of the firm。 While corporate governance best practices ask for the disclosure on the characteristics of the ICS， they do not provide instructions on what management should disclose and on the extent of such disclosure。 Such lack of instructions leaves management with a discretionary choice on the narrative content of ICS disclosure.

This paper offers empirical support for Williamson’s （1983） substitute hypothesis among different governance mechanisms and it has relevant policy implications。 While most corporate governance studies consider disclosure as a complementary mechanism management adopts to reinforce the governance system of the firm （Chen and Jaggi, 2000； Eng and Mak， 2003； Barako et al。, 2006） and indeed provide contrasting results， in this study we show that disclosure on ICS substitutes for other governance mechanisms。 This means that not necessarily better governance implies greater transparency and disclosure。 Firms adhere to corporate governance best practices by disclosing information on the ICS and such disclosure is more extensive when investors need more assurance about the protection of their interests， when other governance mechanisms are weak。 On the other side, when the governance system is sound， management have less incentives to extensively disclose information on the ICS， as this is a costly activity and its benefits are overwhelmed by the other governance mechanisms.

The evidence provided by the empirical research has important policy implications， because it offers insights to firms and practitioners on the relevance of disclosure on internal control systems as a monitoring mechanism for investors。 The remainder of the paper is structured as follows。 The next section reviews the theoretical background and develops the research hypotheses. The research method is

described in section 3， followed by results discussed in section 4。 Concluding remarks are presented in the last section。

Theoretical Background and Hypotheses Development

According to corporate governance literature, the main internal monitoring mechanisms are the board of directors， the ownership structure of the firm, and the internal control system (Gillan， 2006)。 In particular， ICSs play a central role in the protection of investors’ interests both assuring the reliability of financial reporting and promoting the timely identification， assessment and management of relevant risks that encumber upon the business。 The centrality of ICS in corporate governance has been widely recognized by the vast majority of codes of best practice1.

In order to express their concerns and price their claims， investors need to get information on the design and functioning of monitoring mechanisms。 In the cases of mechanisms like the ownership structure and the board of directors, information concerning structure and composition， type and composition of committees in place， number of meetings and so on， is publicly available。 In some other cases, the enforcement of reporting on ICS weaknesses or material deficiencies – like those required by the SOX — provide investors with relevant information about possible gaps in the functioning of the ICS (Leone, 2007）。

Nevertheless， specific information on the characteristics of the ICS is indeed more difficult and expensive to gather because ICSs are complex sets of activities and processes carried out internally to the firm （Deumes and Knechel, 2008； Bronson et al.， 2006）. Indeed, while corporate governance best practices require to disclose information on the ICS, they do not provide instruction on the narrative contents of ICS disclosure. Therefore， investors are unlikely to be informed about the nature, extent, processes and quality of internal controls， unless disclosure on the characteristics of the ICS is provided by the management。 The content and extent of such disclosure will depend on the existing monitoring package （Leftwich et al。, 1981； Williamson, 1983） of the firm。

At the best of our knowledge， disclosure on the specific characteristics and

functioning of ICS has been deserved poor attention。 While the introduction of the SOX in the USA， and the related requirement for disclosure on ICS deficiencies or material weaknesses has increasingly attracted academic interest in recent times （among the others see Ash Baugh et al.， 2007； Doyle et al., 2007； Leone， 2007）, only few studies focused on the specific characteristics of ICS disclosure.

Bronson et al. （2006） examine firm characteristics associated to disclosure on ICS before it was made mandatory by SOX。 They find a positive association between the likelihood of issuing a management report on internal control and corporate governance variables like the number of audit committee meetings and the percentage of institutional shareholders. Deumes and Knechel (2008） identify a list of six disclosure items that capture the ICS information generally available in the annual reports of firms analyzed。 They find that the disclosure index on ICS is significantly associated to variables that proxy for the agency costs of equity and with variables that proxy for agency costs of debt.

According to our theoretical framework, if disclosure on ICS acts as an alternative governance mechanism， when the pricing of claims is high （Jensen and Meckling, 1976） —due to the fact that the other various monitoring devices already in place are not effective enough to limit the costs of the agency relationship — we expect that disclosure on ICS acts as substitute for other monitoring mechanisms in order to reduce the overall intensity of agency conflicts (Williamson， 1983， Fernandez and Arrondo, 2005）.

In order to test this hypothesis， we focus on two fundamental elements of the monitoring package， besides the disclosure on ICS： the ownership structure and the board of directors. Corporate governance studies identify three proxies for the supervisory role of the ownership structure： i） the supervisory role of large investors， ii） the monitoring role of institutional investors and iii） the alignment effect of managerial ownership. We expect that the incentives for management to disclose information on the firm’s ICS will be higher for those firms where the monitoring role played by the owners is weaker。

Literature and empirical evidences attribute to large shareholders a key

supervisory role。 Kang and Shivdasani （1995） detected a positive association between the presence of large shareholders and management's turnover in underperforming firms. On the other side， a disperse ownership is usually associated to a lower monitoring ability and greater information symmetries （Shleifer and Vishny， 1986； Zeckhauser and Pound， 1990; Barako et al. 2006）.Alternatively said, the direct supervision performed by large shareholders reduces the need for alternative monitoring mechanisms。 Consequently， we expect that incentives to disclose on ICS are higher when the ownership is diffused。

Institutional investors also play a relevant supervisory role. While individual investors in public firms have little incentive to monitor management as they are exposed to private costs against which there are public benefits (Grossman and Hart， 1980）, institutional investors have higher incentives to play an active monitoring role on the management because of their large voting power (Milgrom and Roberts， 1992)。 Moreover， institutional investors can access to management through privileged information channels， in order to get disclosure on the firm’s operations （Schadewitz and Blevins， 1998)。 Thus we expect that in presence of institutional investors, management have lower incentives to disclose on ICS.

The last proxy for the supervisory role of the ownership structure is the managerial ownership. It is generally accepted that management’s stock ownership contributes to the alignment of managerial and shareholders' interests (Jensen and Meckling， 1976; Bronson etal.， 2006； Deumes and Knechel， 2008）， thus reducing the agency conflicts inside the firm (Eng and Mak， 2003； Fernandez and Arrondo， 2005 Cheng and Courtenay， 2006)。 As managerial stock ownership reduces the need for monitoring， we expect that incentives to disclose on ICS are higher when the level of managerial ownership is lower。

Boards of directors play a crucial role in monitoring management as shareholders delegate to them the power to control managerial decisions. Previous literature （Carcelo and Neal， 2000；Fernandez and Arrondo， 2005; Krishan， 2005） identifies different proxies for the capability of the board to monitor managerial

behavior ： i） the proportion of independent directors， ii) the presence of CEO duality， iii） the presence of accounting experts and iv) the monitoring ability of the audit committee。 We expect that the more powerful the monitoring role of the board of directors， the lower the incentives for management to disclose information on ICS。 Independent directors are expected to monitor the activities of the board and to limit managerial opportunism (Fama, 1980； Fama and Jensen， 1983)。 Empirical evidences support this expectation。 Rosenstein and Wyatt （1990) explain the positive stock price effects associated to the appointment of a new independent director in terms of positive reaction signals of the markets to the monitoring role played by the outsiders。 A number of studies document a positive relationship between the proportion of independent directors on the board and firms’ performance （Baysinger and Butler， 1985； Goodstein and Boeker， 1991; Pearce and Zahra， 1992）： the proportion of independent directors of the board is considered a proxy of the capability of the board to control managerial actions （Fernandez and Arrondo， 2005） thus supporting a positive association between the proportion of independent members of the board and effectiveness of their monitoring role。 Therefore， we expect that the higher the presence of independent directors， the lower incentives for management to voluntarily disclose on ICS。

—- Sergio Beretta. Disclosure on Internal Control Systems—As a Substitute of Alternative Governance Mechanisms， Bocconi University，Press.2009. 附录：

内部控制系统披露-一种可替代的管理机制

根据代理理论，各种治理机制减少了投资者和管理者之间的代理问题（Jensen and Meckling，1976； Gillan，2006）。传统上，治理机制已经被认定为内部或外部的。内部机制包括董事会及其作用、结构和组成（Fama，1980;Fama and Jensen,1983)，管理股权（Jensen and Meckling，1976）和激励措施，起监督作用的大股东（Demsetz and Lehn，1985),内部控制系统（Bushman and Smith，2001），规章制度和章程条款（反收购措施）和使用的债务融资（杰森，1993)。

外部控制是由公司控制权市场（Grossman and Hart,1980)、劳动力管理市场（Fama，1980）和产品市场(哈特,1983）施加的控制。

各种各样的金融丑闻，动摇了世界各地的投资者，公司治理最佳实践方式特别强调了内部控制系统在公司治理中起到的重要作用。内部控制有助于通过提供保证可靠性的财务报告，和临时议会对可能会损害公司经营目标的事项进行评估和风险管理来保护投资者的利益。这些功能已被的广泛普及内部控制系统架构设计的广泛认可,并指出了内部控制是用以促进效率,减少资产损失风险，帮助保证财务报告的可靠性和对法律法规的遵从（COSO，1992).

尽管有其相关性，但投资者不能直接观察,因此也无法得到内部控制系统设计和发挥功能的信息，因为它们都是组织内的内在机制、活动和过程（Deumes and Knechel,2008）。

由于投资者考虑到成本维持监控管理其声称的（Jensen and Meckling，1976）,内部控制系统在管理激励信息沟通上的特性,以告知投资者内部控制系统的有效性，是当其他监控机制(该公司的股权结构和董事会）比较薄弱,从而为其提供便捷的监控（Leftwich et等, 1981）。存在的替代机制一直是人们在不同公司治理文献中争论的话题（Rediker and Seth, 1995;Fernandez and Arrondo，2005），基于威廉姆森（1983年）的替代假说认为，特定控制机制的边际作用取决于其在公司治理制度的相对重要性。

在本文中，我们认为披露内部控制系统的特点是在管理者选择的监控机制时存在一个可替代治理机制.Leftwich（1981）认为“管理者选择一个监控包,监测包的组成取决于各种监控机制的成本与效益\"(P.59）。

特别是,我们重点关注内部控制系统和监控包的另外两个机制 (公司的所有权结构和董事会） 间的关系，根据有关文献（Jensen and Meckling,1976;Fernandez and Arrondo，2005;Gillan，2006）它们在管理行为监控方面发挥相关作用。我们假设认为，内部控制系统的特性取决于激励由企业的股权结构和董事会发挥监督作用。

因此，我们研究了三年间在四个不同的证券交易所上市（伦敦、巴黎、法兰克福和米兰）160家欧洲公司内部控制系统披露程度,通过利用这个国际范例，我们能够描绘出不同的环境的某些功能。

我们发现证据表明内部控制系统披露代替了监测方面所发挥的作用与所有制结构、制度所有权、在董事会上董事比例和会计审计委员会的专家成员的比率相关.

我们再加上通过内部会计控制的披露所发挥的管理作用，采用以往文献的完整披露架构，使我们能够详细地从内容和信息管理的程度上内制披露上进行交流。而公司治理的最佳做法要求披露内部控制系统的特性，他们没有提供管理应该披露和披露程度的指令.这样的管理缺乏指示使内部控制系统存在随意性。

本文提供了实证支持威廉姆森（1983年成立）在不同的治理机制下的替代假说，它有相应的含义.虽然大多数公司治理研究考虑一个互补的披露机制管理采用强化公司治理系统(Chen and Jaggi，2000;Eng and Mak，2003；Barako 等，2006），的确提供了结果对比，在本研究中,我们表明内部控制系统披露是可替代的其他治理机制。这意味着更好的治理不一定意味着更高的透明度和更多的披露。公司坚持公司治理的最佳做法是提供内部控制系统披露信息,这样当其他治理机制很薄弱可以保证投资者的利益。另一方面，当管理体系健全，管理有更少的激励内部控制体系的广泛披露，因为这是一个高成本的活动,而且它会打压其他治理机制。

其提供的证据实证研究具有重要的含义，因为它提供本研究向企业和从业人员提供披露内部控制系统作为投资者的一种监控机制的相关见解.剩下的论文结构如下。第二节评论的理论背景和发展研究假设。研究方法是第三节中所描述的结果讨论,其次是第四节。作为结束语给出了最后一段.

理论背景和发展

根据相关公司治理文献，主要公司治理的内部监督机制是董事会，公司的所有权结构、内部控制系统（Gillan，2006)。特别是，内部控制系统在保护投资者的利益发挥核心作用:可以促进效率，减少资产损失风险，帮助保证财务报告的可靠性和对法律法规的遵从。内部控制系统在公司治理中扮演的角色已得到广泛认同。

为了表达他们的想法和主张，投资者需要获取监控机制设计与运行的信息.在研究案例的处理机制，如股权结构和董事会有关结构和组成、类型和地方委员会的组成，会议的次数等都是可获取的信息.在其他一些案例中,内部控制系统执

法力度薄弱或不足，如萨班斯法案所要求—向投资者提供内部控制体系可能存在差异的相关信息(Leone，2007）.

不过,对内部控制系统的具体信息的收集的确是困难和昂贵的,因为内部控制体系是在公司中执行的一系列活动和过程（Deumes and Knechel，2008;Bronson等，2006）。实际上,公司治理的最佳做法要求内部控制系统的披露，但它们不提供内部控制系统具体内容的披露。因此，投资者也不太可能了解其性质、范围、程序和内部控制质量，除非内部控制系统披露报告是由管理者提供的。这样的披露内容和程度取决于公司现行的监控包(Leftwich 等,1981；Williamson,1983)。

据我们所知,内部控制系统的特点和功能的披露还没有引起广泛的注意.美国萨班斯法案的出台,内部控制系统按相关要求披露弱点和不足在近几年日益受到学术兴趣（among the others see Ash Baugh 等人,2007;Doyle 等人，2007；Leone,2007)，只有少数研究焦点在于内部控制披露方面。

Bronson等人，（2006） 探讨有关公司的内部控制系统特性的披露它是由萨班斯法案强制执行的。他们发现了一个积极的可能性之间的关联发布了内部控制管理报告、公司的治理变数如审计委员会会议的次数、机构股东的百分比.Deumes and Knechel （2008)确定了内部控制系统信息披露可以从公司年度报告中获得的六个披露项目的清单.他们发现,内部控制披露指数，显示了相关变量代表的股权代理成本和负债代理成本变量。

根据我们的理论框架，如果内部控制系统披露作为一个替代治理机制，当索赔定价索赔高(Jensen and Meckling， 1976)由于这一事实，其他已经实行各种监测设备没有足够有效的代理关系的成本,我们期望内部控制披露，以作为其他监督机制替代，以减少代理冲突的整体强度（Williamson，1983； Fernandez and Arrondo，2005)。

为了检验该假说，我们关注的两个基本要素的监控方案，除了内部控制系统披露,还包括股权结构和董事会。公司治理研究识别三个代表监督作用的所有权结构包括：（1）大型投资者的监督作用；（2）机构投资者的监督作用；（3）经理层持股对齐的效果.我们期望管理的激励机制对公司内部控制系统的披露会被这些监察作用发挥较弱的公司使用。

文献与实证指出了关键的大股东所起的监督作用。Kang and Shivdasani （1995）检测到发现了大股东之间的存在比率和管理层在业绩不佳的公司营业额呈正相关。另一方面,一个分散的所有权通常关联一个到较低的监测能力和更多的信息对称性(Shleifer and Vishny， 1986；Zeckhauser and Pound,1990；Barako 等，2006），或者说，由公司的大股东直接监督,减少了对替代监控机制的需要。因此,我们希望鼓励内部控制系统披露的公司的所有权是分散的。

机构投资者也起到了有关监督作用。虽然上市公司的个人投资者很少有动力去监督管理,他们很容易手私人成本与公共利益的影响(Grossman and Hart，1980)，机构投资者所持有的投票权激励他们的监督管理作用（Milgrom and Roberts，1992）。此外，机构投资者可以通过信息渠道特权参与管理，以取得该公司的运营披露(Schadewitz and Blevins， 1998）。因此,我们预期机构投资者的存在,管理有较低的内部控制系统披露.

最后代表监督作用的所有权结构就是管理所有权。人们普遍认为管理股权有助于对齐的管理人员和股东的利益（Jensen and Meckling,1976;Bronson 等， 2006； Deumes and Knechel, 2008),因而减少了在公司代理冲突（Eng and Mak, 2003； Fernandez and Arrondo, 2005 Cheng and Courtenay，2006).管理股权减少了监控的需要，我们希望鼓励内部控制系统披露是在管理所有权较低的情况下。

董事会在监控管理作为股东委托给他们的权力控制的管理决策的重要作用.以往文献（Carcelo and Neal,2000；Fernandez and Arrondo，2005； Krishan，2005)确定了董事委员会的能力，进行不同的代理管理行为的监察：(1)董事比例；(2)CEO的二元性存在;（3）会计专家的比率;（4）审计委员会的监控能力。我们期望的更强大董事会监控作用、较低内部控制系统披露.预计董事、董事会的监察活动，可以经理的机会主义（Fama,1980； Fama and Jensen，1983)。实证证据支持了这个预期。Rosenstein and Wyatt (1990）解释正相关的股票价格与一个新的董事任命和外部人员在市场的监督作用方面发挥了积极的反应信号相关联。大量的研究证实董事比例对公司的董事会和公司绩效的一个正面的关系 （Baysinger and Butler，1985； Goodstein and Boeker,1991； Pearce and Zahra,1992)：董事在董事会的比率被认为是以控制管理代理的行动能力（Fernandez and Arrondo，2005） 从而证明了的董事会和他们的监督作用的有效性的成员的比例呈正相关.因此，我们期望更高董事的存在，降低激励管理层自愿披露内部控制系统.