切勿相信这七条关于董事会的讹传

许多公司治理专家非常关注与董事会相关的问题，因为董事会需要监督一家企业的方方面面（如战略、资本结构、风险、绩效等），肩负着招聘和辞退CEO之责。此外，当企业运营出现问题时，董事会需要向股东及时解释。

由于董事会具有如此重要的地位，人们都希望企业在挑选董事会成员时能够遵循一些最佳实践，其中一些是监管机构和股票交易所的强制要求，还有一些是来自各路专家的建议。

但这些“最佳实践”未必总能提高董事会的效率或质量，现在就让我们揭穿7个所谓“最佳实践”的真相。

讹传1：董事长应该始终独立

企业治理领域流传最广的信条之一，就是一家企业的CEO不应兼任董事长。过去10年里，许多股东要求标普500指数成分股公司剥离这两个职位，此类委托提案高达300多份。迪士尼、摩根大通和美国银行等知名企业都收到过类似提案。

许多企业都遵从了这种要求。2014年，在标普500指数成分股公司中，CEO兼任董事长的企业只占53%，较2005年的71%大幅降低。在同一时间段内，由完全独立的外部人士出任董事长的企业则由9%上升至28%。

尽管有些人认为，保持独立有利于董事长对企业和管理层进行更严格的监督，然而研究证据并不支持这一结论。一项研究发现，董事长是否独立与企业运营绩效并无统计学关联。另一项研究则表明，没有证据显示这种独立地位的改变（不管是让两个职位合并还是剥离）会影响企业未来的运营绩效。有研究甚至发现，强迫剥离董事长与CEO职权，对企业的运营绩效实际是有害的。那些迫于投资者压力而拆分了这两项职权的企业，往往会在宣布决定的日期前后得到负面回报，运营绩效也会随之降低。

讹传2：分期分级董事会制度必然不利于股东

很多人认为，分期分级董事会制度（又称董事轮换制）使管理层感受不到市场压力，因而不利于股东。在分期分级董事会架构下，董事会实际上是三年一选而不是一年一选，因为每年选举的董事只占董事会的三分之一。由于大多数董事无法在一年内替换，收购者也就无法在短期内搞定多数董事。所以说，分期分级董事会架构实际上形成了一种强大的反收购保护。正因为如此，许多企业管理专家都在批评这种制度。

过去10年内，采取分期分级董事会制度的企业大大减少，从2005年的57%下降到了2014年的32%。大盘股是取消这种董事会制度比例最大的公司类型。诚然，分期分级董事会在个别情况下不利于股东，比如有可能阻碍非常划算的收购机会，或是巩固表现不佳的管理层，然而在多数情况下，这种董事会架构却能够改善企业绩效。比如，它能逼退恶意收购者，从而确保企业的长期战略不受干扰；再比如，它能够保护管理层承担较少的短期压力，从而使企业敢于创新和承担风险，研发暂时还不被市场理解的专有技术。

一项研究显示，分期分级董事会制度有助于改善新上市企业的长期运营绩效。此外还有研究表明，该架构有助于企业管理层放手进行长期投资，最终让公司斩获最大利益。

讹传3：达到纽交所独立标准的股东才算真正独立

仅仅由于一名董事满足纽交所的独立董事标准，并不意味着向企业管理层提供建议或履行监督之责时，他或她就会保持真正独立的态度。

在2009年的一项研究中，研究者分别对符合纽交所标准的独立董事（形式独立）和那些在教育、经验、提拔等方面与CEO没有关联的董事（社交独立）进行了调查。结果发现，与CEO有社交关系的董事，不管是否符合纽交所的独董标准，都有可能会过于信任或依赖CEO，从而在行事上有失偏颇。另外他们更有可能向CEO支付更多薪水，同时更不愿意辞退运营绩效不佳的CEO。

其他一些研究也给出了类似的结论。比如一份研究发现，由CEO指定的董事更可能认同CEO的决定，因此他们的独立度更低。在现任CEO任期内任命的董事成员比例越大，董事会监管职能的效果就越差。虽然独立是外部董事的一个重要品质，但纽交所的标准未必能真正衡量出一名董事究竟独立与否。

讹传4：连锁董事会降低公司治理质量

所谓“连锁董事”（Interlocked directorship），是指A公司某位高管担任B公司的董事，B公司某位高管又担任了A公司的董事。企业管理专家经常批评连锁董事现象会产生“你好我好大家好”的心理，从而削弱董事会的独立性和监督职能。虽然有些证据显示，连锁董事的确会造成这种现象，但也有研究表明，连锁董事也可以有利于股东。

连锁董事可以在董事之间建立起一张网络，从而促进信息，以及战略、运营和监督最佳实践在相关公司之间流通。另外，连锁董事的网络效应也可以作为企业重要的业务关系渠道，为企业介绍新客户和供应商，并有助于企业获得人才、资本和政治关系。

有研究表明，连锁董事的网络效应提高了风投行业的绩效。同时，如果几家企业在高管层和董事层分享了关系网，那么他们的投资政策往往更趋于近似，盈利水平也更高。还有一些研究发现，董事会与董事会之间的连锁关系会带来更成功的并购、更高的运营绩效以及更高的股价回报。

讹传5：CEO是最好的董事

许多专家都相信CEO是最好的董事，因为他们的管理知识使他们能为企业各方面的监督指导提供广泛的帮助，比如在战略、风险管理、接班人计划、绩效评估、股东与利益相关者关系等方面。股东们往往也相信这一点，因此每当有CEO被任命为董事，股东的反应都比较积极。但从经验证据上看，这个结论还有待商榷。

研究发现，没有证据能够表明CEO董事能为企业的远期运营绩效或决策做出更突出的贡献。研究还发现，CEO董事往往倾向于给CEO开更高的薪水。另外，海德思哲公司和斯坦福大学岩石企业治理中心联合进行的一项调查表明，大多数的董事都认为，在职CEO往往忙于自己公司的事，没法成为一名有效的董事会成员。过去15年里，在企业新招募的独董中，在职CEO的比重已经有所下降。很多企业都在招募低于CEO级别的高管或退休CEO担任董事。

讹传6：董事承担着重大的债务风险

三分之二的董事认为，近年来董事面临的债务风险呈上升势头，有15%的董事因担心该职务有可能带来个人债务而曾经认真考虑过辞职。不过，实际上，由董事掏私人腰包偿还公司债务的风险还是比较低的。通过董事补偿协议和董事与高管责任保险等措施，董事们实际上被给予了相当程度的保护。董事补偿协议规定，当企业遭到股东集体诉讼，或陷入与信托义务有关的案件时，只要董事确系秉诚行事，企业就得补偿董事的损失。董事与高管责任保险又加了一层额外保护，该保险的赔付范围包括诉讼费用、调解费用和一定限额内的损失赔偿。事实证明，这些手段能够有效保护董事个人为企业“背锅”。

一项研究显示，在1980年到2005年的25年间，外部董事个人为企业承担债务的案例（即没有获得补偿和保险赔偿）仅有12起。后续研究发现，在2006年至2010年间的所有诉讼中，没有一位外部董事自掏腰包为企业“背锅”（虽然目前有些案件仍在审理）。研究人员总结道：“在最新的保险政策保护下，董事个人承担债务的风险是很低的。对于表现不佳的外部董事来说，最大的威胁是诉讼带来的时间成本、事态的进一度恶化，以及对名誉的潜在影响，而非直接的经济损失。”

讹传7：公司的失败总是董事会的错

为了让公司产生可以接受的回报率，企业必须要承担风险，而风险偶尔也会导致失败。如果失败是由于战略不成熟、风险承担得太多、监督疲软或公然欺诈导致的，那么我们可以，也应该将失败的责任归咎于董事会。但如果失败是由于竞争压力、市场的意外变动导致的，甚至如果我们原本就预计到了可能出现较差的结果，那么责任就在管理层身上，或者纯粹是运气不佳。

即便就董事会的监督职能而言，董事会也未必能发现每一起渎职事件的征兆。董事会掌握的企业运营信息是有限的，它主要依赖管理层提供的信息来了解他们的决策。除了少数情况下，董事会一般不会“越界”寻找信息（比如除非收到了举报，或是基于对管理层语言和行动的观察产生了一种不祥之感。）

然而，有证据表明，董事会还是会因为亏损而受到惩罚。2005年的一项研究表明，在公司更正财务报表后，董事会成员换人的几率会显著增加，夸大收益的企业的董事也容易被踢出董事会。与之类似，在金融危机期间担任大型金融机构（如美国银行、美林、摩根斯坦利、美联、华盛顿互惠银行等）的董事，也成为了旨在将他们踢出董事会的“反对票”运动的目标。

要想确定一名董事到底要承担何种程度的责任，就应当公正地评估企业的失败究竟是不是由他导致的，究竟是如何导致的，以及是否有理由相信他曾经设法阻止不良后果的发生。

本文作者David F. Larcker是斯坦福大学会计学教授，Brian Tayan是斯坦福商学院研究员。（财富中文网）

译者：朴成奎

审校：任文科

Corporate governance experts pay considerable attention to issues involving boards of directors, and with good reason. Boards are responsible for monitoring all aspects of a business (its strategy, capital structure, risk, and performance), hiring and firing the CEO, and answering to shareholders when something goes awry.

Because of the importance of these roles, companies are expected to adhere to best practices, some mandated by regulatory standards and stock exchange requirements and some advocated by experts.

But these “best practices” don’t always create better board effectiveness or quality. Here we debunk seven of these practices.

Myth 1: The chairman should always be independent.

One of the most widely held beliefs in corporate governance is that the CEO of a company should not serve as its chairman. In fact, over the last decade, companies in the S&P 500 Index received more than 300 shareholder-sponsored proxy proposals that would require a separation of the two roles. Shareholder groups have targeted prominent corporations including Walt Disney, JP Morgan, and Bank of America to strip their CEOs of the chairman title.

Companies, in turn, have moved toward separating the roles. Only 53% of companies in the S&P 500 Index had a dual chairman/CEO in 2014, down from 71% in 2005. Similarly, the prevalence of a fully independent chair increased from 9% to 28% over this period.

Despite the belief that an independent chair provides more vigilant oversight of the organization and management, the research evidence does not support this conclusion.One study found no statistical relationship between the independence status of the chairman and operating performance, while another found no evidence that a change in independence status (separation or combination) impacts future operating performance. Additional research actually found that forced separation is detrimentalto firm outcomes: Companies that separate the roles due to investor pressure exhibit negative returns around the announcement date and lower subsequent operating performance.

Myth 2: Staggered boards are always detrimental to shareholders.

Many believe that staggered boards harm shareholders by insulating management from market pressure. Under a staggered board structure, directors are elected to three-year rather than one-year terms, with one-third of the board standing for election each year. Because a majority of the board cannot be replaced in a single year, staggered boards are a formidable antitakeover protection, and for this reason many governance experts criticize their use. Over the last 10 years, the prevalence of staggered boards has decreased, from 57% of companies in 2005 to 32% in 2014. The largest decline has occurred among large capitalization stocks. While it is true that staggered boards can be detrimental to shareholders in certain settings — such as when they prevent otherwise attractive merger opportunities and entrench a poorly performing management — in other settings they have been shown to improve corporate outcomes. For example, they benefit shareholders when they protect long-term business commitments that would be disrupted by a hostile takeover or when they insulate management from short-term pressure, thereby allowing a company to innovate, take risk, and develop proprietary technology that is not fully understood by the market. One study found staggered boards improve long-term operating performance among newly public companies. Other studies also suggest that staggered boards can benefit companies by committing management to longer investment horizons.

Myth 3: Directors who meet NYSE independence standards are independent.

Just because a director satisfies the independence standards of the New York Stock Exchange does not mean he or she behaves independently when it comes to advising and monitoring management. For example, a 2009 study examined directors who are independent according to NYSE standards (“conventionally independent”) and those who are independent in their social relation to the CEO based on education, experiences, and upbringing (“socially independent”). The researchers discovered that board members who share social connections can be biased to overly trust or rely on CEOs, regardless of whether they’re considered independent by NYSE standards. Those board members were more likely to pay CEOs more and less likely to fire a CEO following poor operating performance.

Other studies reach similar conclusions. One study found that directors appointed by a CEO are more likely to be sympathetic to his or her decisions and therefore less independent. The greater the percentage of the board appointed during the current CEO’s tenure, the worse the board performs its monitoring function. While independence is an important quality for an outside director to have, NYSE standards do not necessarily measure its presence (or absence).

Myth 4: Interlocked directorships reduce governance quality.

Interlocked directorships occur when an executive of Firm A sits on the board of Firm B while an executive of Firm B sits on the board of Firm A. Corporate governance experts criticize board interlocks as creating psychological reciprocity that compromises independence and weakens oversight. While some evidence suggests that interlocking can create this effect, research also suggests that interlocking can be beneficial to shareholders. Interlocking creates a network among directors that can lead to increased information flow, whereby best practices in strategy, operations, and oversight are transferred across companies. Network effects created by interlocked directorships can also serve as an important conduit for business relations, client and supplier referrals, talent sourcing, capital, and political connections. For example, one study found thatnetwork connections improved performance among companies in the venture capital industry, while another found that companies that share network connections at the senior executive and the director level have greater similarity in their investment policies and higher profitability. These effects disappear when network connections are terminated. Other studies have found board connections lead to more successful mergers and acquisitions, and greater future operating performance and higher future stock price returns.

Myth 5: CEOs make the best directors.

Many experts believe that CEOs are the best directors because their managerial knowledge allows them to contribute broadly to firm oversight, including strategy, risk management, succession planning, performance measurement, and shareholder and stakeholder relations. Shareholders, too, often have this belief, reacting favorably to the appointment of CEOs to the board. But empirical evidence is less positive. Studies have found no evidence that a CEO board member positively contributes to future operating performance or decision-making and finds CEO directors are associated with higher CEO pay. Additionally, a survey by Heidrick& Struggles and the Rock Center for Corporate Governance at Stanford University finds that most corporate directors believe that active CEOs are too busy with their own companies to be effective board members. Over the last 15 years, the percentage of newly recruited independent directors with active CEO experience has declined. Companies instead are recruiting new directors who are executives below the CEO level or who are retired CEOs.

Myth 6: Directors face significant liability risk.

Two-thirds of directors believe that the liability risk of serving on boards has increased in recent years, and 15 percent of directors have thought seriously about resigning due to concerns about personal liability. However, the actual risk of out-of-pocket payment is low. Directors are afforded considerable protection through indemnification agreements and director and officer liability insurance. Indemnification agreements stipulate that the company will pay for costs associated with securities class actions and fiduciary duty cases, provided the director acted in good faith. Insurance provides an additional layer of protection, covering litigation expenses, settlement payments, and, in some cases, amounts paid in damages up to a specified limit. These protections have been shown to be effective in protecting directors from personal liability. One study found that in the 25 years between 1980 and 2005, outside directors made out-of-pocket payments — meaning unindemnified and uninsured — in only 12 cases. A follow-up study of lawsuits filed between 2006 and 2010 finds no cases resulting in out-of-pocket payments by outside directors (although some of these cases are still ongoing). The authors conclude that “directors with state-of-the art insurance policies face little out-of-pocket liability risk. … The principal threats to outside directors who perform poorly are the time, aggravation, and potential harm to reputation that a lawsuit can entail, not direct financial loss.”

Myth 7: The failure of a company is always the board’s fault.

In order for a company to generate acceptable rates of returns, it must takes risks, and risks periodically lead to failure. If the failure was the result of a poorly conceived strategy, excessive risk taking, weak oversight, or blatant fraud, the board can and should be blamed. But if failure resulted from competitive pressure, unexpected shifts in the marketplace, or even poor results that fall within the range of expected outcomes, then blame lies with management or poor luck.

Even within the scope of its monitoring obligations, a board won’t necessarily detect all instances of malfeasance before they occur. The board has limited access to information about the operations of a company. In the absence of “red flags,” it relies on the information provided by management to inform its decisions. A board usually doesn’t seek information beyond this except in a few cases (if it receives whistleblower information, for example, or believes management isn’t setting the right tone through words or behavior).

Still, evidence shows boards are punished for losses. A 2005 study showed that director turnover increases significantly following financial restatements and that board members of firms that overstate earnings tend to lose their other directorships as well. Similarly, directors who served on the boards of large financial institutions during the financial crisis (such as Bank of America, Merrill Lynch, Morgan Stanley, Wachovia, and Washington Mutual) became targets of “vote no” campaigns to remove them from other corporate boards where they served.

The degree to which a director should be held accountable depends on a fair-minded assessment of whether and how the director might have contributed to the failure and whether it is reasonable to believe that he or she could have prevented it.

David F. Larcker is the James Irvin Miller Professor of Accounting and Brian Tayan is a researcher at Stanford GSB.